Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Science and Technology 588
_____________________________ _____________________________
Synthesis, Properties and Applications of Chalcogen-Containing Antioxidants
BY
JONAS MALMSTRÖM
ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000
Dissertation for the Degree of Doctor of Philosophy in Organic Chemistry Presented atUppsala University in 2000
ABSTRACTMalmström, J. 2000. Syntheses, Properties, and Applications of Chalcogen-ContainingAntioxidants. Acta Univ. Ups. Comprehensive Summaries of Uppsala Dissertations from theFaculty of Science and Technology 588. 64 pp. Uppsala. ISBN 91-554-4863-1.
This thesis deals with the synthesis, properties, and applications of chalcogen-containingantioxidants. In the first part, the preparation and properties of chalcogen-containing vitamin E analoguesare described. The sulfur compound 3,3,4,6,7-pentamethyl-2,3-dihydrobenzo[b]thiophene-5-ol was prepared by two different routes using ionic and radical chemistry. Interestingrearrangements were observed in the two synthetic pathways. A new methodology for the synthesis of dihydroselenophene and dihydrotellurophenederivatives is described. In the preparation of the vitamin E analogues 2,3-dihydrobenzo[b]selenophene-5-ol and 2,3-dihydrobenzo[b]tellurophene-5-ol a tellurium-mediated tandem SRN1/SHi sequence was suggested to be operative. 2,3-Dihydrobenzo[b]thiophene-5-ol and the vitamin E-like selenide 2-methyl-2-(4,8,12-trimethyl-tridecyl)-selenochroman-6-ol were prepared via intramolecular homolytic substitution atsulfur and selenium, respectively. The first rate constant for intramolecular homolyticsubstitution at tellurium is also reported (5x108 s-1 at 25 °C). The antioxidant profile for 2,3-dihydrobenzo[b]furan-5-ol and its 1-thio, 1-seleno, and 1-telluro analogues is described. By means of pulse radiolysis, it was shown that the one-electron reduction potentials (ArO·/ArO-) were independent of the chalcogen (0.49-0.52 V vsNHE). The O-H bond dissociation enthalpies for the compounds were also estimated to besimilar (336-340 kJ mol-1). The pKa values and the oxidation potentials were also determinedfor these compounds. For some compounds the rate of hydrogen atom donation to tert-butoxyl radicals was determined by means of laser flash photolysis. Using a two-phase lipidperoxidation model, it was demonstrated that the selenium and tellurium analogues could beregenerated in the presence of a stoichiometric amount of a reducing agent. Theorganotellurium analogue also acted as a good glutathione-peroxidase mimic and as a potentinhibitor of lipid peroxidation in liver microsomes. In the second part of the thesis the stabilizing capacity of bis[4-(dimethylamino)phenyl]telluride was investigated in the thermoplastic elastomer PACREL®.It was demonstrated that the addition of 0.17-0.50 % of the telluride significantly improvedthe tensile strength and elongation at break of the polymer. Chemiluminescencemeasurements showed that the organotellurium compound prolonged the induction period ofthermo-oxidation and reduced the total luminescence intensity of the material.
Jonas Malmström, Department of Organic Chemistry, Institute of Chemistry, University ofUppsala, Box 531, SE-751 21 Uppsala, Sweden
© Jonas Malmström 2000
ISSN 1104-232XISBN 91-554-4863-1
Printed in Sweden by Eklundshofs Grafiska AB, Uppsala 2000
Det finns inga omöjliga drömmar - bara vår begränsade uppfattning om vad som är möjligt.Beth Mende Conny
Till mina föräldrar och Ulrika
Papers included in this thesis
This thesis is based on the following papers and appendix, referred to in the text by theirRoman numerals.
I. Novel Antioxidants: Unexpected Rearrangements in the RadicalCyclization Approach to 2,3-Dihydrobenzo[b]thiophene-5-ol Derivatives.Malmström, J; Gupta, V.; Engman, L. J. Org. Chem. 1998, 63, 3318-3323. Copyright ©1998, American Chemical Society.
II. Toward Novel Antioxidants: Preparation of Dihydrotellurophenes andSelenophenes by Alkyltelluride-Mediated Tandem SRN1/SHi Reactions.Engman, L.; Laws, M. J.; Malmström, J.; Schiesser, C. H.; Zugaro L. M.J. Org. Chem. 1999, 64, 6764-6770. Copyright © 1999, American Chemical Society.
III. The Antioxidant Profile of 2,3-Dihydrobenzo[b]furan-5-ol and its 1-Thio, 1-Selenoand 1-Telluro Analogues.Malmström, J.; Jonsson, M.; Cotgreave, I. A.; Hammarström, L.; Sjödin, M.; Engman,L. J. Am. Chem. Soc. 2000, Submitted for publication. Copyright © 2000, AmericanChemical Society.
IV. Synthesis of Novel Selenium-Containing Vitamin E Analogues via IntramolecularHomolytic Substitution (SHi) at Selenium.
Malmström, J.; Engman, L.; Schiesser, C. H.Preliminary Manuscript.
V. Stabilization of PACREL by Organotellurium Compound.Malmström, J.; Engman, L.; Bellander, M.; Jacobsson, K.; Stenberg, B.; Lönnberg, V.J. Appl. Polym. Sci. 1998, 70, 449-456. Copyright © 1998, John Wiley & Sons, Inc.
VI. Appendix: Supplementary material.Malmström, J.
Reprints were made with kind permission from the publishers.
CONTENTS
Abstract
Papers included in this thesis
List of abbreviations
1. INTRODUCTION ............................................................................................................ 1
1.1 Oxygen in biological and polymeric systems ......................................................... 11.1.1 General introduction ........................................................................................ 1
1.1.2 Mechanism of autoxidation .............................................................................. 1
1.1.3 Endogenous defense systems ............................................................................ 21.1.4 Oxidative degradation and stabilization of polymers ...................................... 3
1.2 Selenium and tellurium in organic chemistry ........................................................ 41.2.1 Introduction ...................................................................................................... 4
1.2.2 Nucleophilic selenium and tellurium and preparation of diselenides andditellurides 5
1.2.3 Other applications ............................................................................................ 6
1.3 Radical cyclizations in organic synthesis ............................................................... 71.3.1 Introduction ...................................................................................................... 7
1.3.2 Methods to conduct radical cyclizations .......................................................... 81.3.3 Intramolecular homolytic substitution (SHi)..................................................... 11
2. CHALCOGEN ANALOGUES OF VITAMIN E .......................................................... 13
2.1 Introduction .............................................................................................................. 132.2 Synthesis of chalcogen-containing vitamin E analogues ...................................... 16
2.2.1 Preparation of a sulfur analogue of vitamin EI ................................................ 16
2.2.2 Towards selenium and tellurium analogues of vitamin EII............................... 232.2.3. Synthesis of 2,3-dihydrobenzo[b]furan-5-ol and its 1-thio analogueIII........... 32
2.3 The antioxidant profile of 2,3-Dihydrobenzo[b]furan-5-ol and its 1-thio,1-seleno and 1-telluro analoguesIII................................................................................. 33
2.3.1 Introduction ...................................................................................................... 33
2.3.2 Redox and thermochemical properties ............................................................. 33
2.3.3 Reactivity towards tert-butoxyl radicals .......................................................... 352.3.4 Inhibition of lipid peroxidation ........................................................................ 36
2.3.5 Hydroperoxide decomposing capacity ............................................................. 392.3.6 Lipid peroxidation in liver microsomes ........................................................... 41
2.4 Towards a selenium-containing tocopherolIV......................................................... 422.4.1 Introduction ...................................................................................................... 422.4.2 Strategy ............................................................................................................. 42
2.4.3 Synthesis of the model compound 2,2-dibutyl selenacyclohexane ................... 43
2.4.4 Synthesis of a selenium-containing tocopherol analogue ................................ 45
2.5 Conclusions and outlook .......................................................................................... 46
3. ORGANOTELLURIUM COMPOUNDS AS STABILIZERS FORPOLYMERIC MATERIALSV ............................................................................................ 48
3.1 Introduction .............................................................................................................. 48
3.2 Preparation of bis[4-(dimethylamino)phenyl] telluride ........................................ 49
3.3 Autoxidation and mechanical properties in polymers .......................................... 493.3.1 Chemiluminescence .......................................................................................... 49
3.3.2 Mechanical properties ...................................................................................... 50
3.4 Results and discussion .............................................................................................. 513.4.1 Mechanical properties ...................................................................................... 51
3.4.2 Chemiluminescence .......................................................................................... 54
3.4.3 Dynamic mechanical properties ....................................................................... 553.4.4 Stabilizer stability ............................................................................................. 56
3.5 Conclusions and outlook ................................................................................... 56
5. ACKNOWLEDGEMENTS ............................................................................................. 57
List of abbreviations
Ac AcetylADP Adenosine 5’-diphosphateAIBN α,α´-AzobisisobutyronitrileAMVN 2,2’-Azobis(2,4-dimethylvaleronitrile)Ar ArylBDE Bond Dissociation EnthalpyBHT 3,5-Di-t-butyl-4-hydroxytolueneBn BenzylBu ButylCB-A Chain-breaking acceptingCB-D Chain-breaking donatingCL ChemiluminescenceDMAP 4-DimethylaminopyridineDMF DimethylformamideDMSO Dimethyl sulfoxideEt EthylFc FerroceneGSH GlutathioneGSH-px Glutathione peroxidaseGSSG Glutathione disulfideHPLC High Performance Liquid ChromatographyIn InitiatorINEPT Insensitive Nuclei Enhanced by Polarization TransferLFP Laser flash photolysism-CPBA meta-Chloroperbenzoic acidMD Machine directionMe MethylMs MethanesulfonylNAC N-acetylcysteineNADPH Nicotinamide adenine dinucleotide phosphateNHE Normal Hydrogen ElectrodeNMR Nuclear Magnetic ResonanceNOE Nuclear Overhauser EnhancementPD Peroxide decomposerPh PhenylPTOC Pyridine-2-thione-N-oxycarbonylrt Room temperatureSOMO Singly Occupied Molecular OrbitalTBAF Tetrabutylammonium fluorideTBAP Tetrabutylammonium perchlorateTBARS Thiobarbituric acid-reactive substancesTBDMS tert-ButyldimethylsilylTD Transverse directionTEMPO 2,2,6,6-Tetramethyl-1-piperidinyloxy, free radicalTHF TetrahydrofuranTLI Total Luminescence IntensityTs ToluenesulfonylTTMSS Tris(trimethylsilyl)silane
1
1. Introduction
1.1 Oxygen in biological and polymeric systems
1.1.1 General introduction
Molecular oxygen is essential to the energy producing processes of aerobic organisms,
but it is also detrimental in many cases.1 All organic materials exposed to molecular
oxygen undergo oxidative degradation (autoxidation). This process can be delayed by
the involvement of antioxidants.2 An antioxidant is generally defined as an inhibitor
that is effective in preventing oxidation by molecular oxygen.
Historically, de Saussure was the first to study autoxidation phenomena when in 1820
he measured the rate of walnut oil oxidation. A few years later, Berzelius suggested
that the autoxidation of oil was initiated by atmospheric oxygen. In the 1920�s,
Moureau and Dufraisse pioneered the study of antioxidants and the first approved
antioxidant for food came in 1933. Today, antioxidants have great commercial interest
in the preservation of various foods and in the prevention of oxidative degradation of
petroleum products, rubber, and plastics. Medicinally, they are also interesting in the
prevention of various diseases.3,4
1.1.2 Mechanism of autoxidation
The autoxidation process involves a radical chain reaction initiated by environmental
stresses such as heat, irradiation or metal ions. Scott has proposed a general
mechanism for autoxidation, consisting of two interlinked processes (Scheme 1).5 An
alkyl radical, R·, formed as a result of environmental stress, is a common intermediate
in the two cycles. The alkyl radical readily reacts with molecular oxygen to form the
harmful peroxyl radicals, ROO·,6 which in turn can abstract a hydrogen atom, thus
regenerating alkyl radicals and producing hydroperoxides. Hydroperoxides can easily
1 Ernster, L. Chem. Scr. 1986, 26, 525.2 See for example: (a) Ingold, K. U. Chem. Rev. 1961, 61, 563. (b) Denisov, E. T.; Khudyakov, I. V.
Chem. Rev. 1987, 87, 1313. (c) Packer, L.; Cadenas, E., Eds., Antioxidants in Health and Disease,Vol. 3. Handbook of Antioxidants and Vol. 4. Handbook of Synthetic Antioxidants, Marcel Dekker:New York, 1997.
3 Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and Medicine, Oxford University Press:Oxford, 1999.
4 For a review on pharmaceutical antioxidants, see Andersson, C.-M.; Hallberg, A.; Högberg, T. Adv.Drug Res. 1996, 28, 67.
5 Scott, G. Chem. Brit. 1985, 21, 648.6 Ingold, K. U. Acc. Chem. Res. 1969, 2, 1.
2
produce the reactive hydroxyl radicals (HO·) and alkoxyl radicals (RO·) in the
presence of light, heat or a transition metal. Unless the autoxidation is terminated in
some way, for example by combination of two radicals or by the involvement of
antioxidants, the chain reaction will continue. Fortunately, aerobic organisms have
developed several endogenous defense systems against the over-production of reactive
oxygen species (oxidative stress, see Section 1.1.3).
CHAIN
BREAKING
Chain-breaking donating antioxidants
Chain-breaking accepting antioxidants
Environmentalstress Preventive antioxidants
Environmentalstress
RH
HO+RO
RHROOH
O2
ROOR
Chain-breaking donatingantioxidants
Scheme 1. General mechanism for autoxidation.
1.1.3 Endogenous defense systems
Reactive oxygen species (for example ROO·, HO·, O2-· and H2O2) are produced
continuously in our body as a consequence of normal metabolic processes as well as
from environmental stresses. These species need to be inactivated, otherwise they
might cause various diseases, including atherosclerosis, cancer, and an accelerated
aging process.3,7 For example, lipids containing polyunsaturated fatty acids and their
esters are readily oxidized by molecular oxygen in a process called lipid
peroxidation.8 Normally, however, any over-production of reactive oxygen species
can be controlled by enzymatic and non-enzymatic antioxidants. Antioxidants can be
divided into two groups depending on their mode of action. Preventive antioxidants
act primarily through attenuating the initiation step whereas chain-breaking
7 Sies, H. Angew. Chem., Int. Ed. Engl. 1986, 25, 1058.8 (a) Niki, E. Chem. Phys. Lipids 1987, 44, 227. (b) Gardner, H. W. Free Rad. Biol. Med. 1989, 7,
65.
3
antioxidants act through termination of the free radical chain reaction (Scheme 1).
Peroxide decomposers (PD), UV absorbers and metal chelators belong to the group of
preventive antioxidants. In biological systems, the enzymes catalase and the
glutathione peroxidases (GSH-px) represent this kind of antioxidant. The chain-
breaking antioxidants can be divided into two groups depending on their mechanism
of action, chain-breaking donating antioxidants (CB-D) and chain-breaking accepting
antioxidants (CB-A). Typical examples of CB-D type antioxidants are sterically
hindered phenols, e.g. BHT (1), and aromatic amines, e.g. diphenylamine (2), which
can donate a hydrogen atom. Nitroxyl radicals, e.g. TEMPO (3), which can trap short-
lived free radicals by combining with them, are examples of the accepting type of
antioxidants.
OH
1
NH
2
NO
3
Unfortunately, our natural defense against reactive oxygen species is not perfect. It is
therefore of great interest to design and synthesize new antioxidants with higher
antioxidative capacity than the naturally occurring ones.
1.1.4 Oxidative degradation and stabilization of polymers9
Polymers are also susceptible to attack by reactive oxygen species. Exposure to
oxygen, ozone, high temperature, UV light, radiation, and chemical agents can
degrade them. The thermo- and photo-oxidative aging of polymers causes a
deterioration in their physical and mechanical properties. Poorer mechanical strength
often results and properties such as stiffness, creep resistance, and brittleness are
affected.
The differences between aging processes are often related to the nature of the
initiating step of a free-radical chain reaction in which molecular oxygen is one of the
9 (a) Zaikov, G. E.; Ya Polishchuk, A. Russ. Chem. Rev. 1993, 62, 603. (b) Shlyapnikov, Y. A.;
Kiryushkin, S. G.; Mar�in, A. P. Antioxidative Stabilization of Polymers, Taylor and Francis:London, 1996. (c) Zweifel, H. Stabilization of Polymeric Materials, Springer-Verlag: Berlin, 1998.
4
reactants. Polyethylene for example, is degraded 105 times more rapidly in oxygen
atmosphere than in nitrogen at 260 °C.10 The autoxidation of simple low molecular
weight compounds is relatively well understood, but the oxidation of polymers is
more complex.11
The free-radical chain reaction is initiated by thermal, photochemical or mechanical
strain or by decomposition of peroxides or other impurities incorporated into the
polymer during processing. The deleterious consequences of oxidative aging can be
relieved by the addition of chain-breaking antioxidants, UV-absorbers, metal-
complexing agents, and peroxide decomposers. However, commercial stabilizers
sometimes suffer from poor efficiency and toxicity, so there is a great need for better
antioxidants/stabilizers for polymeric materials.
1.2 Selenium and tellurium in organic chemistry
1.2.1 Introduction
Berzelius discovered elemental selenium in 1818 and the first organoselenium
compounds were prepared a few years later by Löwig and Wöhler. These compounds
were simple aliphatic selenols (RSeH), selenides (RSeR), and diselenides (RSeSeR).
Von Reichenstein isolated elemental tellurium in 1782 and Wöhler prepared the first
organotellurium compound, Et2Te, in 1840. Although the fields of organoselenium
and -tellurium chemistry date back to the 19th century, it was not until the 1970�s
when selenium and tellurium came into practical use in organic chemistry. The
apparent turning point was Sharpless use of the phenyl selenide anion as a mild
reagent for converting epoxides into allylic alcohols (vide infra).12 Ever since, there
has been a growing interest in the development of organoselenium and -tellurium
chemistry.
A full overview of the use of selenium and tellurium in organic chemistry will not be
given here. A number of review articles13 and books14 have been published on this
topic.
10 Turi, E. A. Thermal Characterization of Polymeric Materials, Academic Press: NY, 1981; p. 848.11 Allen, N. S. Degradation and Stabilization of Polymers, Applied Science Publishers: London, 1983.12 Sharpless, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1973, 95, 2697.13 Selenium: (a) Reich, H. J. Acc. Chem. Res. 1979, 12, 22. (b) Liotta, D. Acc. Chem. Res. 1984, 17,
28. (c) Clive, D. L. J. Tetrahedron 1978, 34, 1049. Tellurium: (d) Engman, L. Acc. Chem. Res.1985, 18, 274. (e) Petragnani, N.; Comasseto, J. V. Synthesis 1991, 793. (f) Idem. ibid. 1991, 897.(g) Comasseto, J. V.; Barrientos-Astigarraga, R. E. Aldrichim. Acta 2000, 33, 66.
5
1.2.2 Nucleophilic selenium and tellurium and preparation of diselenides and
ditellurides
Due to their importance in the synthetic work in this thesis, nucleophilic selenium and
tellurium species and the preparation of diselenides and ditellurides will be discussed
in some detail.
Selenide or telluride anions are often used to prepare organoselenium and
organotellurium compounds. They are very powerful, soft nucleophiles, which react in
an SN2 fashion with a wide variety of substrates, including alkyl halides and epoxides.
For example, Sharpless used the phenyl selenide anion to ring open the epoxide 4 to
the alcohol 5 in excellent yield (Scheme 2).12 The anions can be formed by reduction
of diselenides or ditellurides, respectively. The reduction is commonly carried out
using NaBH4 in ethanol, but other reductants can also be used.14a,15
HOH2O2
SePh
HOO
SePh
EtOH2 h, rt.
4 5
Scheme 2. Ring opening of an epoxide by the strongly nucleophilic phenyl selenide anion.
The corresponding diorganyl diselenides and ditellurides can be prepared in various
ways. Diselenides are generally prepared, as depicted in Scheme 3, by oxidation of
selenols or selenolates,16 or by the reaction of Li2Se2 or Na2Se2 with alkyl or aryl
halides. Aryl diselenides can also be obtained from the reaction of Na2Se2 with
diazonium salts.
14 For example (a) Nicolaou, K. C.; Petasis, N. A. Selenium in Natural Product Synthesis, CIS,
Philadelphia, PA, 1984. (b) Back, T., Ed., Organoselenium Chemistry – A Practical Approach,Oxford University Press: Oxford, 1999. (c) Irgolic, K. J. In �Houben-Weyl Methoden derOrganischen Chemie�, Organotellurium Compounds, 4th edn., Klamann, D., Ed., Thieme: Stuttgart,1990; Vol. E12b. (d) Petragnani, N.; Tellurium in Organic Synthesis; Academic Press: London,1994.
15 (a) Salmond, W. G.; Barta, M. A.; Cain, A. M.; Sobala, M. C. Tetrahedron Lett. 1977, 1683. (b)Reich, H. J.; Shah, S. K. J. Org. Chem. 1977, 42, 1773.
16 Selenolates and tellurolates are most commonly prepared by the reaction of Grignard reagents oralkyl- or aryl lithiums with elemental selenium or tellurium.
6
RSe-
Na2Se2 + 2 RX
RN2+X- + Na2Se2
R=aryl
RSeSeR
RMgX or RLi + Se ox.
Scheme 3. General methods for the preparation of diselenides.
Ditellurides are generally prepared, as outlined in Scheme 4, by oxidation of
tellurolate anion,16 alkylation or arylation of Na2Te2, or by reduction of the
corresponding organyl tellurium trichlorides.
ox.RMgX or RLi + Te
RTeTeR
RTeCl3
Na2Te2 + 2 RX
RTe-
red.
Scheme 4. General methods for the preparation of ditellurides.
1.2.3 Other applications
Organoselenium and -tellurium compounds have also found several other applications
in chemistry other than the purely synthetic uses. For example, they can be used as
building blocks for organic conductors.17 Organoselenium compounds also play
important biological roles. Thus, selenocysteine residues are incorporated in the
glutathione peroxidase enzymes, which serve to reduce hydroperoxides without
formation of toxic radical intermediates. Diaryl chalcogenides are a stable and
odorless class of compounds, which have attracted a lot of interest in this research
laboratory. It has recently been found that these compounds can act as selective
inhibitors of thioredoxin reductase, and, thereby, act as potential antitumor agents.18
17 (a) Cowan, D.; Kini, A. in The Chemistry of Organic Selenium and Tellurium Compounds, Patai, S.
and Rappoport, Z., Eds., John Wiley & Sons: Chichester, 1987; Vol. 2, Chap. 12. (b) Wudl, F. InOrganoselenium Chemistry, Liotta, D., Ed.; John Wiley & Sons: New York, 1987; Chap. 9. (c)Smith, L. M.; Thompson, J. Chemtronics 1989, 4, 60. (d) McCullough, R. D.; Kok, G. B.; Lerstrup,K. A.; Cowan, D. O. J. Am. Chem. Soc. 1987, 109, 4115.
18 Engman, L.; Cotgreave, I.; Angulo, M.; Taylor, C. W.; Paine-Murrieta, G. D.; Powis, G. AnticancerRes. 1997, 17, 4599.
7
They can also protect against peroxynitrite-mediated oxidation in our body.19 Diaryl
tellurides also serve as effective antioxidants in polymeric systems.20,V We have also
demonstrated that various selenium and tellurium-containing compounds can act as
chain-breaking antioxidants or peroxide decomposers in model systems.21,22
Apparently, new organoselenium and organotellurium compounds could find many
uses.
1.3 Radical cyclizations in organic synthesis
1.3.1 Introduction
The development of radical chemistry started in the early twentieth century when
Gomberg investigated the formation and reactions of the triphenylmethyl radical.23 In
the 1920�s Paneth showed that less stabilized radicals also exist and he measured their
lifetime in the gas phase.24 The use of radicals in organic synthesis began in the
1930�s when Kharasch recognized that the anti-Markovnikov addition of hydrogen
bromide to alkenes proceeds via a radical chain process.25 At about the same time,
Hey and Waters described the phenylation of aromatic compounds with benzoyl
peroxide as a radical reaction.26
The use of free radicals in organic synthesis has increased dramatically during the last
twenty years, and radical-based methodology has become an important tool in organic
19 (a) Briviba, K.; Tamler, R.; Klotz, L.-O.; Engman, L.; Cotgreave, I. A.; Sies, H. Biochem.
Pharmacol. 1998, 55, 817. (b) Jacob, C.; Arteel, G. E.; Kanda, T.; Engman, L.; Sies, H. Chem. Res.Toxicol. 2000, 13, 3.
20 Engman, L.; Stern, D.; Stenberg, B. J. Appl. Polym. Sci. 1996, 59, 1365.21 (a) Cotgreave, I. A.; Moldéus, P.; Engman, L.; Hallberg, A. Biochem. Pharmacol. 1991, 42, 1481.
(b) Andersson, C.-M.; Brattsand, R.; Hallberg, A.; Engman, L.; Persson, J.; Moldéus, P.; Cotgreave,I. A. Free Rad. Res. 1994, 20, 401. (c) Engman, L.; Persson, J.; Vessman, K.; Ekström, M.;Berglund, M.; Andersson, C.-M. Free Rad. Biol. Med. 1995, 19, 441. (d) Vessman, K.; Ekström,M.; Berglund, M.; Andersson, C.-M.; Engman, L. J. Org. Chem. 1995, 60, 4461. (e) Wieslander, E.;Engman, L.; Svensjö, E.; Erlandsson, M.; Johansson, U.; Linden, M.; Andersson, C.-M.; Brattsand,R. Biochem. Pharmacol. 1998, 55, 573. (f) Kanda, T.; Engman, L.; Cotgreave, I. A.; Powis, G. J.Org. Chem. 1999, 64, 8161.
22 (a) Cotgreave, I. A.; Moldéus, P.; Brattsand, R.; Hallberg, A.; Andersson, C.-M.; Engman, L.Biochem. Pharmacol. 1992, 43, 793. (b) Engman, L; Stern, D.; Cotgreave, I. A.; Andersson, C.-M.J. Am. Chem. Soc. 1992, 114, 9737. (c) Andersson, C.-M.; Hallberg, A.; Brattsand, R.; Cotgreave, I.A.; Engman, L.; Persson, J. Bioorg. Med. Chem. Lett. 1993, 3, 2553. (d) Engman, L.; Stern, D.;Pelcman, M.; Andersson, C.-M. J. Org. Chem. 1994, 59, 1973. (e) Engman, L.; Andersson, C.;Morgenstern, R.; Cotgreave, I. A.; Andersson, C.-M.; Hallberg, A. Tetrahedron 1994, 50, 2929.
23 (a) Gomberg, M. Ber. Dtsch. Chem. Ges. 1900, 33, 3150. (b) Gomberg, M. J. Am. Chem. Soc.1900, 22, 757.
24 Paneth, F.; Hofeditz, W. Ber. Dtsch. Chem. Ges. 1929, 62, 1335.25 Kharasch, M. S.; Margolis, E. T.; Mayo, F. R. J. Org. Chem. 1936, 1, 393.26 Hey, D. H.; Waters, W. A. Chem. Rev. 1937, 21, 169.
8
synthesis,27 especially for natural product synthesis.27d The advantages of radical
reactions are their high functional group tolerance and the use of mild reaction
conditions combined with high levels of regio- and stereoselectivity. Radical
cyclization reactions are a powerful and versatile method for the construction of
mono- and polycyclic systems.28 This can be exemplified by the elegant one-pot
synthesis of the taxane skeleton (6) reported by Pattenden (Scheme 5).29
O
O
I
n-Bu3SnHH
H
O
OH
6
Scheme 5
1.3.2 Methods to conduct radical cyclizations
”Metal hydride” methods
Tri-n-butyltin hydride30 and tris(trimethylsilyl)silane31 (TTMSS) are commonly used
reagents for effecting reductive radical carbon-carbon bond formation. The
mechanism for radical cyclization using tin hydride as the promoter can be divided
into three steps: initiation, propagation and termination, as exemplified in Figure 1.
An analogous chain can be written for TTMSS. Initiation of the radical chain is
usually accomplished by thermal decomposition of α,α´-azobisisobutyronitrile
(AIBN) to generate a tin radical. When lower temperatures are needed, light or Et3B in
the presence of trace amounts of oxygen,32 are commonly used. In the propagating
step, the initial radical 8 is generated from a radical precursor 7 by atom or group
transfer to the tin radical formed in the initiation step. The radical 8 can then undergo
the desired intramolecular addition to the double bond to form the radical 9, which
27 For selected reviews, see: (a) Curran, D. P. Synthesis 1988, 417. (b) Idem. ibid. 1988, 489. (c) Hart,
D. J. Science 1984, 223, 883. (d) Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev. 1991, 91,1237. (e) Ramaiah, M. Tetrahedron 1987, 43, 3541.
28 Giese, B; Kopping, B.; Göbel, T.; Dickhaut, J.; Thoma, G.; Kulicke, K. J.; Trach, F. OrganicReactions 1996, 48, 301.
29 Hitchcock, S. A.; Pattenden, G. Tetrahedron Lett. 1992, 33, 4843. Corrigendum see: Idem. ibid.1992, 33, 7448.
30 Neumann, W. P. Synthesis 1987, 665.31 Chatgilialoglu, C. Acc. Chem. Res. 1992, 25, 188.32 (a) Miura, K.; Ichinose, Y.; Nozaki, K.; Fugami, K.; Oshima, K.; Utimoto, K. Bull. Chem. Soc. Jpn.
1989, 62, 143. (b) Nozaki, K.; Oshima, K.; Utimoto, K. J. Am. Chem. Soc. 1987, 109, 2547.
9
finally undergoes an intermolecular reaction with the tin hydride to produce the
cyclized product 10. A competing reaction is the reduction of radical 8 to form the
reduced compound 11. If cyclization is slow it is necessary to use low concentrations
of the hydride to minimize the undesired process. In this case it is also advisable to
use TTMSS as a hydride donor, since it is about 10 times less reactive than n-Bu3SnH
toward alkyl radicals. There are also some other practical problems associated with
the use of tin hydride.31,33 Tin reagents are toxic and they often make product isolation
difficult.34 Therefore, a wide range of methods has been developed to avoid the use of
tin.35 The most notable method is the Barton method, which will be discussed in the
next section. A wide range of other methods has also been used for conducting radical
cyclizations.28
Termination: Disproportionation, reaction with solvent, or oxidation.
+ In-HR3SnInInitiation: R3Sn-H
Propagation:
9 8
7
R3SnH
R3SnX
X
H
R3Sn
11
10
R3SnH
Figure 1. Chain mechanism for radical cyclization mediated by trialkylstannyl radicals.
Many radical precursors, including halogens, phenyl selenides or -sulfides, and
xanthate esters, can be used. The reactivity of the carbon heteroatom bond (RX)
towards stannyl radicals is generally in the order I > Br > SePh ≈ OC(S)SMe > Cl >
33 Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 7200.34 The desired product is often contaminated with tin compounds after flash chromatography.35 Baguley, P. A.; Walton, J. C. Angew. Chem., Int. Ed. Engl. 1998, 37, 3072.
10
SPh.36,37 The order of reactivity of various R· toward tin hydrides is aryl ≈ vinyl >
alkyl > allyl ≈ benzyl.
Intramolecular cyclization to a double bond can occur in two different ways - exo or
endo (Figure 2). Generally, the exo mode of cyclization is kinetically preferred.
C C
X
kexo
kendo
exo
endo
C
X
C
CX
C
Figure 2. Exo and endo modes of cyclization.
Radical additions to C=C double bonds are usually irreversible and exothermic, with
early, reactant-like transition states. Such kinetically controlled reactions often give
rise to products that are unavailable by conventional ionic methods.27d
The free-radical methodology is most successful for the synthesis of 5-membered
rings. There are several reasons for this: 1) The rate of cyclization for the formation of
5-membered rings is usually higher than the cyclization to give any other ring size.
For example, the simple 5-hexenyl radical cyclizes 20 times faster than does the 6-
heptenyl radical.38 2) The regioselectivity for 5-exo cyclizations is often
outstanding.38,39 3) Cyclization giving 5-membered rings can be highly
stereoselective.38,40
The thiohydroxamate method (Barton method)
The Barton41 method is one of the most important radical chain methods where tin is
not involved (Scheme 6). This reaction provides a convenient way to generate alkyl
36 Beckwith, A. L. J.; Pigou, P. E. Aust. J. Chem. 1986, 39, 77.37 Ingold, K. U.; Lusztyk, J.; Scaiano, J. C. J. Am. Chem. Soc. 1984, 106, 343.38 (a) Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron Lett. 1985, 26, 373. (b) Beckwith, A. L. J.;
Schiesser, C. H. Tetrahedron 1985, 41, 3925. (c) Spellmeyer, D. C.; Houk, K. N. J. Org. Chem.1987, 52, 959.
39 Surzur, J. M. In Reactive Intermediates, Abramovitch, R. A., Ed.; Plenum Press: New York, 1980;Vol. 2, Chap. 3.
40 (a) RajanBabu, T. V.; Fukunaga, T. J. Am. Chem. Soc. 1989, 111, 296. (b) RajanBabu, T. V.;Fukunaga, T.; Reddy, G. S. J. Am. Chem. Soc. 1989, 111, 1759. (c) Beckwith, A. L. J.; Easton, C.J.; Lawrence, T.; Serelis, A. K. Aust. J. Chem. 1983, 36, 545.
41 For a general review on Barton chemistry, see: Barton, D. H. R. Aldrichim. Acta 1990, 23, 3.
11
radicals from carboxylic acids. The radical chain is initiated by light or heat induced
decomposition of the radical precursor thiohydroxamate ester 12,42 which is normally
prepared in situ from carboxylic acids.43 The propagation steps involve cyclization,
addition of the cyclized radical to the thiohydroxamate moiety, and fragmentation
with the expulsion of carbon dioxide to generate the initial radical and the product
alkyl pyridyl sulfide 13.44 In contrast to the metal hydride based reactions, the chain-
transfer step is not hydrogen abstraction but transfer of a thiopyridyl group that can be
removed or employed for further transformations. In addition, if a suitable radical trap
is added, which can trap R2·, then a wide variety of compounds can be prepared.
N
SOR1
O
NR2S
cyclization
12
hν or ∆
Initiation:
Propagation:
12
N
SOR1
O+ +
+ + N
S
13
R1 CO2
R1 R2
- CO2 R1R2
Scheme 6. Schematic representation of the Barton method.
The Barton method can be extended to involve PTOC oxalates (anhydrides of an
oxalic acid monoester and N-hydroxypyridine-thione), which can be used to generate
alkyl radicals from tertiary alcohols. This will be discussed in section 2.4.2.
1.3.3 Intramolecular homolytic substitution (SHi)45
Generally, homolytic substitution results in the displacement of one radical by another
according to equation 1.
42 Barton, D. H. R.; Crich, D.; Potier, P. Tetrahedron Lett. 1985, 26, 5943.43 (a) Barton, D. H. R.; Samadi, M. Tetrahedron 1992, 48, 7083. (b) Barton, D. H. R.; Hervé, Y.;
Potier, P.; Thierry, J. Tetrahedron 1987, 43, 4297. (c) Idem. ibid. 1988, 44, 5479. (d) Barton, D. H.R.; Ozbalik, N.; Vacher, B. Tetrahedron 1988, 44, 3501.
44 Barton, D. H. R.; Bridon, D.; Fernandez-Picot, E.; Zard, S. Z. Tetrahedron 1987, 43, 2733.45 For a general review on homolytic substitution, see: Walton, J. C. Acc. Chem. Res. 1998, 31, 99.
12
R + AB RA + B (1)
The reaction may either proceed via an SH2 mechanism or, more rarely, via an SH1
mechanism. The SH2 reaction is a bimolecular process where R· attacks AB with
subsequent expulsion of B·. Most commonly, SH2 reactions involve attack on a
univalent atom such as hydrogen or a halogen, but it can also occur at multivalent
atoms. If the reaction is intramolecular it is denoted SHi and it can be used to obtain
various heterocycles.46 The reaction has frequently been used for the preparation of
sulfur heterocycles,47 but it has recently been applied to the synthesis of selenium48
and tellurium49 heterocycles as well. For example, irradiation of radical precursor 14
leads to rapid and efficient intramolecular homolytic substitution at selenium to give
substituted and saturated selenium-containing heterocycles 15 in good yield (Scheme
7).48a,b Examples have also been reported where intramolecular homolytic substitution
take place at oxygen,50 carbon,51 silicon,52 and boron53.
Ph Se
R'
OR
O
N
S
hνn nPh Se
R'
RSe
R'Rn
14 15
Scheme 7-
46 Schiesser, C. H.; Wild, L. M. Tetrahedron 1996, 52, 13265.47 See, for example (a) Kampmeier, J. A.; Evans, T. R. J. Am. Chem. Soc. 1966, 88, 4096. (b) Benati,
L.; Montevecchi, P. C.; Tundo, A.; Zanardi, G. J. Chem. Soc., Perkin Trans. 1 1974, 1272. (c)Beckwith, A. L. J.; Boate, D. R. J. Chem. Soc., Chem. Commun. 1986, 189. (d) Beckwith, A. L. J.;Boate, D. R. J. Org. Chem. 1988, 53, 4339. (e) Beckwith, A. L. J. Chem. Soc. Rev. 1993, 143. (f)Crich, D.; Yao, Q. J. Org. Chem. 1996, 61, 3566. (g) Crich, D.; Hao, X. J. Org. Chem. 1997, 62,5982.
48 (a) Schiesser, C. H.; Sutej, K. J. Chem. Soc., Chem. Commun. 1992, 57. (b) Schiesser, C. H.; Sutej,K. Tetrahedron Lett. 1992, 33, 5137. (c) Benjamin, L. J.; Schiesser, C. H.; Sutej, K. Tetrahedron1993, 49, 2557. (d) Lyons, J. E.; Schiesser, C. H.; Sutej, K. J. Org. Chem. 1993, 58, 5632. (e) Fong,M. C.; Schiesser, C. H. J. Org. Chem. 1997, 62, 3103. (f) Lucas, M. A.; Schiesser, C. H. J. Org.Chem. 1998, 63, 3032. (g) Schiesser, C. H.; Zheng, S.-L. Tetrahedron Lett. 1999, 40, 5095. (h)Lucas, M. A.; Nguyen, O. T. K.; Schiesser, C. H.; Zheng, S.-L. Tetrahedron 2000, 56, 3995.
49 Laws, M. J.; Schiesser, C. H. Tetrahedron Lett. 1997, 38, 8429.50 For example: Colombani, D.; Maillard, B. J. Org. Chem. 1994, 59, 4765.51 Ashcraft, M. R.; Buny, A.; Cooksey, C. J.; Davies, A. G.; Gupta, B. D.; Johnson, M. D.; Morris, H.
J. Organomet. Chem. 1980, 195, 89.52 (a) Kulicke, K. J.; Chatgilialoglu, C.; Kopping, B.; Giese, B. Helv. Chim. Acta 1992, 75, 935. (b)
Studer, A. Angew. Chem., Int. Ed. Engl. 1998, 37, 462. (c) Studer, A.; Steen, H. Chem. Eur. J.1999, 5, 759.
53 Batey, R. A.; Smil, D. V. Angew. Chem., Int. Ed. Engl. 1999, 38, 1798.
13
2. Chalcogen Analogues of Vitamin E
2.1 Introduction
Vitamin E is a well-known lipid-soluble antioxidant in biological systems.54 The term
vitamin E refers to the structurally related phenolic compounds called tocopherols
(compounds 16a-d), where α-tocopherol (16a) is the most active one.
R1
R2
HO
XR3
16a X= O, R1=R2=R3= CH3 (α-tocopherol)16b X= O, R1=R3= CH3, R2=H (β-tocopherol)16c X= O, R2=R3= CH3, R1= H (γ-tocopherol)16d X= O, R1=R2= H, R3= CH3 (δ-tocopherol)16e X= S, R1=R2=R3= CH3
The tocopherols protect cell membranes from oxidative degradation by acting as
chain-breaking donating antioxidants. Being phenolic compounds, they can act by
trapping two peroxyl radicals according to equations 2 and 3.
ROO + ArOH ROOH + ArO
ROO + ArO Non-radical combination products
(2)
(3)
Since the pioneering work on vitamin E analogues by Ingold and co-workers in the
1980s,55 there has been a continuous search for compounds with a better antioxidative
capacity than α-tocopherol.56 For example, Ingold55c and Barclay57 prepared oxygen
54 Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194.55 (a) Burton, G. W.; Ingold, K. U. J. Am. Chem. Soc. 1981, 103, 6472. (b) Barclay, L. R. C.; Ingold,
K. U. J. Am. Chem. Soc. 1981, 103, 6478. (c) Burton, G. W.; Doba, T.; Gabe, E. J.; Hughes, L.;Lee, F. L.; Prasad, L.; Ingold, K. U. J. Am. Chem. Soc. 1985, 107, 7053. (d) Robillard, B.; Ingold,K. U. Tetrahedron Lett. 1986, 27, 2817. (e) Robillard, B.; Hughes, L.; Slaby, M.; Lindsay, D. A.;Ingold, K. U. J. Org. Chem. 1986, 51, 1700. (f) Zahalka, H. A.; Robillard, B.; Hughes, L.; Lusztyk,J.; Burton, G. W.; Janzen, E. G.; Kotake, Y.; Ingold, K. U. J. Org. Chem. 1988, 53, 3739.
56 See, for example: (a) Mukai, K.; Okabe, K.; Hosose, H. J. Org. Chem. 1989, 54, 557. (b) Gilbert, J.C.; Pinto, M. J. Org. Chem. 1992, 57, 5271. (c) Mukai, K.; Kageyama, Y.; Ishida, T.; Fukuda, K. J.Org. Chem. 1989, 54, 552.
57 (a) Barclay, L. R. C.;Vinqvist, M. R.; Mukai, K.; Itoh, S.; Morimoto, H. J. Org. Chem. 1993, 58,7416. (b) Barclay, L. R. C.; Edwards, C. D.; Mukai, K.; Egawa, Y.; Nishi, T. J. Org. Chem. 1995,60, 2739.
14
analogues 17 and 18, respectively. These analogues exhibited better antioxidant
capacity than α-tocopherol, mainly due to steric and stereoelectronic effects.55c,57a
Recently, Niki and co-workers prepared the tocopherol analogue 19 as a drug for
inhibition of lipid peroxidation in vivo.58 Certain requirements should be fulfilled for a
good chain-breaking tocopherol: 1) the aromatic ring should be fully methylated; 2)
the overlap between the phenoxyl SOMO and a lone-pair on the chromane oxygen
should be as good as possible. Thus, reducing the ring size of the non-aromatic ring
from six atoms in α-tocopherol to five atoms in the dihydrobenzofuran analogue
increases the activity.54
O
HO
O
HO
O
HO
17 18 19
Although a wide range of lipophilic as well as hydrophilic59 chromanol systems have
been studied, little effort has been made to change the heteroatom in the fused
heterocyclic ring to see how that will affect the overall antioxidative capacity of the
compounds. It is known that the sulfur atom is more effective than oxygen at
stabilizing a neighbouring radical center.60 However, Ingold and co-workers have
prepared 1-thio-α-tocopherol (16e) and related 6-hydroxythiochromanes, and
demonstrated that they were slightly less efficient than the structurally related 6-
hydroxychromanes.55d-f It has been refered that some phenolic compounds carrying
sulfur para to the hydroxylic group, have shown high antioxidant capacity.61 As
mentioned in Section 1.2.3, we have demonstrated that various selenium and
tellurium-containing compounds show antioxidative properties in biological and
58 Watanabe, A.; Noguchi, N.; Fujisawa, A.; Kodama, T.; Tamura, K.; Cynshi, O.; Niki, E. J. Am.
Chem. Soc. 2000, 122, 5438.59 For example, see: (a) Bolkenius, F. N.; Grisar, J. M.; De Jong, W. Free Rad. Res. Commun. 1991,
14, 363. (b) Grisar, J. M.; Petty, M. A.; Bolkenius, F. N.; Dow, J.; Wagner, J.; Wagner, E. R.;Haegele, K. D.; De Jong, W. J. Med. Chem. 1991, 34, 257.
60 See for example: (a) Biddles, I.; Hudson, A.; Wiffen, J. T. Tetrahedron 1972, 28, 867. (b) Griller,D.; Nonhebel, D. C.; Walton, J. C. J. Chem. Soc., Perkin Trans. 2 1984, 1817. (c) Luedtke, A. E.;Timberlake, J. W. J. Org. Chem. 1985, 50, 268.
15
polymeric systems.20,21 For example, bis(4-hydroxyphenyl) telluride (20) was
demonstrated to be a better inhibitor of azo-initiated peroxidation of linoleic acid than
its corresponding sulfur analogue.21d It is also known that tellurides react readily with
hydroperoxides to form tellurium (IV) dihydroxides according to Scheme 8.21d,f These
species can be reduced back to the divalent state by mild stoichiometric reductants,
such as thiols or ascorbate. In this way, the compound can act as a catalyst for
reduction of hydroperoxides.
RSSR 2 RSH
H2O2 or ROOH
R Te ROH
OHR Te R
Scheme 8. Proposed catalytic mechanism for the hydroperoxide
decomposing action of diorganyl tellurides in the presence of thiols.
With this background we thought it would be of interest to synthesize chalcogen-
containing vitamin E analogues, and to investigate if the selenium and tellurium-
containing analogues, in addition to their chain-breaking properties, could be
regenerated according to the mechanism depicted in Scheme 8.
61 (a) Nishiyama, T.; Kagimasa, T.; Yamada, F. Can. J. Chem. 1994, 72, 1412. (b) Yamamura, T.;
Tanaka, K.; Tomiyama, S.; Sai, S.; Yamada, F.; Nishiyama, T. J. Am. Oil Chem. Soc. 1995, 72,1047.
16
2.2 Synthesis of chalcogen-containing vitamin E analogues
2.2.1 Preparation of a sulfur analogue of vitamin EI
The target molecules of interest are the chalcogen-containing vitamin E analogues
21a-c, where the heteroatom X could be sulfur, selenium, or tellurium. Initially it was
decided to synthesize 3,3,4,6,7-pentamethyl-2,3-dihydrobenzo[b]thiophene-5-ol (21a).
Te OHHOX
HO
20 21a X=S21b X=Se21c X=Te
It is known that the 2,3-dihydrobenzothiophene as well as the 2H-1-benzothiopyran
systems can be made by cyclodehydration of the corresponding aryl hydroxyalkyl
sulfides.62 Preparation of the alcohol 25 required for such an approach was completed
in a couple of steps starting from commercially available 2,3,6-trimethylphenol (22)
(Scheme 9). Bromination of phenol 22 with bromine in acetic acid occurred,
regiospecifically,63 in the 4-position to give the bromophenol 23, which was then
protected as its TBDMS ether 24 in excellent yield. After protection, the bromide 24
was treated with tert-butyllithium, elemental sulfur and isobutylene oxide in a one-pot
procedure to obtain the desired alcohol 25 in an isolated yield of 74 %. Finally, the
alcohol was treated with neat concentrated sulfuric acid to obtain an inseparable 1:1
mixture of the desired product 21a and the by-product 26, in low yield. The low yield
may be explained as a consequence of steric interactions between the gem-dimethyl
group and the methyl group in the 4-position.
62 See, for example: (a) El-Khawaga, A. M.; El-Zohry, M. F.; Ismail, M. T.; Abdel-Wahab, A. A.;
Khalaf, A. A. Phosphorus Sulfur 1987, 29, 265. (b) Spruce, L. W.; Gale, J. B.; Berlin, K. D.;Verma, A. K.; Breitman, T. R.; Ji, X.; van der Helm, D. J. Med. Chem. 1991, 34, 430.
63 Mikhailov, V. A.; Savelova, V. A.; Rodygin, M. Y. Russ. J. Org. Chem. 1993, 29, 1868.
17
HO HO
Br
aTBDMSO
Br
b c,d,e
TBDMSO
S
OHf
HO
S+
HO
S
22 23 24
21a 2625Ratio 1:1
Scheme 9. (a) Br2, AcOH, 20 °C, 97 %. (b) TBDMSCl, imidazole, DMF, 20 °C, 95 %. (c) t-BuLi,THF, -78 °C. (d) Sulfur, THF, 20 °C. (e) Isobutylene oxide, THF, 20 °C, 74 % from 24. (f) Conc.H2SO4, 20 °C, 27 %.
To obtain separation of sulfides 21a and 26 by flash chromatography for
characterization, the compounds were oxidized to their respective sulfoxides by m-
CPBA.
Cyclodehydration of the alcohol 25 was found to be unsuccessful in polyphosphoric,
perchloric, and formic acid. Various Lewis acids also failed to cause cyclization using
either nitromethane or toluene as solvents.
The formation of the dibenzo[b]thiophene derivative 26 is mechanistically interesting.
It could be proposed to occur via the carbocation and episulfonium ion intermediates,
as outlined in Scheme 10, followed by both methyl and hydride shifts and finally
oxidation. The driving force behind this rearrangement would be the aromatization of
the molecule.
HO
S
HO
S
HO
S
H--shift HO
S
HO
S
HO
S
[ox]HO
S
26
Scheme 10. Proposed mechanism for formation of the dibenzo[b]thiophene derivative 26.
18
Since ionic chemistry did not work as well as expected, this led to the consideration of
various radical cyclizations for the preparation of sulfide 21a. Cyclization of an aryl
radical could be a convenient way for obtaining the target molecule. The radical
precursors 28 and 29 were prepared as outlined in Scheme 11. Bromination of the
alcohol 25 resulted not only in aromatic bromination, but partial substitution of the
tertiary alcohol as well. The crude bromide was therefore hydrolyzed to afford the
brominated alcohol 27 in a 67 % isolated yield from the alcohol 25. The acid-
catalyzed dehydration of compound 27 afforded the terminal olefin 28 as the kinetic
product and the vinylic sulfide 29 as the thermodynamic one, in good yields.
a,bTBDMSO
S
BrOH
TBDMSO
S
Br
TBDMSO
S
Br
c
d
28
29
2725
TBDMSO
S
OH
Scheme 11. (a) Br2, AcOH, 20 °C. (b) Et3N, THF, H2O, 67 % from 25. (c) p-TsOH, toluene, reflux, 1h, 68 %. (d) p-TsOH, toluene, reflux 24 h, 74 %.
The radical precursors were then subjected to radical cyclization (Scheme 12). Thus,
the terminal olefin 28 was treated with tri-n-butyltin hydride and a catalytic amount of
AIBN to afford a 3:1 mixture of the desired compound 30 and its constitutional
isomer 31, in moderate yield. The reaction was also carried out using an excess of the
hydride donor. With a 5-fold excess, compound 30 was isolated as the sole cyclized
product in low (17 %) yield, probably due to suppression of the 6-endo mode of
cyclization. In the cyclization of vinylic sulfide 29, the poorer hydride donor
tris(trimethylsilyl)silane was added by syringe pump together with AIBN to keep the
formation of reduced starting material 32 to a minimum. In addition to reduced
starting material 32 (50 %), a 1:1 mixture of the desired compound 30 and isomer 31
was isolated in 41 % yield.
19
30
TBDMSO
S
31
TBDMSO
S+28 29
Ratios a 30:31 3:1
a b
b 30:31 1:1
Scheme 12. (a) n-Bu3SnH, AIBN, C6H6, reflux, 3.5 h, 40 %. (b) TTMSS, AIBN, C6H6,
reflux, slow addition of reagents during 8 h followed by reflux for 15 h, 41 %.
TBDMSO
S
H
32
It was possible to obtain the pure sulfide 30 by careful flash chromatography of the
mixture, but unfortunately it was not possible to isolate pure compound 31. In order to
characterize compound 31, the mixture was treated with m-CPBA to afford the
corresponding sulfoxides. Compound 33 was then separable by flash chromatography.
As indicated in Figure 3, the structural assignments were supported by NOE-
difference experiments and were further corroborated by selective INEPT
experiments.
S
t-BuSi
30
S
H Ht-BuSi
O33
Figure 3. Nuclear Overhauser Effects for compounds 30 and 33.
Finally, sulfide 30 was deprotected with tetra-n-butylammonium fluoride (TBAF) to
obtain the desired product 21a in 79 % isolated yield (Scheme 13).
20
TBDMSO
S
30
aHO
S
21a
Scheme 13. (a) TBAF, THF, 20 °C, 79 %.
Despite the low yields in the cyclization reactions, the mechanisms for formation of
the isomer 31 are interesting. The proposed mechanisms are outlined in Schemes 14
and 15. It could be proposed that the radical 34 can either undergo a 5-exo cyclization,
followed by hydrogen abstraction to form the desired compound 30, or it can undergo
a 6-endo cyclization to form the 6-membered radical 35. This species can then
undergo β-scission to form the sulfur-centered radical 36. Subsequent 5-exo
cyclization, followed by hydrogen abstraction, would then give the isomer 31.
Interestingly, the ratio between the two isomers was independent of the tin hydride
concentration used (0.35 or 0.05 M). It is therefore unlikely that a neophyl
rearrangement is involved in this reaction.64 Unless electron-withdrawing substituents
are present in the aromatic ring, the neophyl rearrangement is normally slow.64a
TBDMSO
S
5-exo [H]
6-endo
TBDMSO
S
1) 5-exo
2) [H]β-scission
34 30
3136
TBDMSO
S
TBDMSO
S
TBDMSO
S
35
TBDMSO
S
Scheme 14. Proposed mechanisms for the formation of compounds 30 and 31 from radical 34.
The analogous mechanisms for formation of sulfides 30 and 31 from precursor 37 are
more speculative (Scheme 15). It may be considered that the radical 37 can either
64 (a) Abeywickrema, A. N.; Beckwith, A. L. J.; Gerba, S. J. Org. Chem. 1987, 52, 4072. (b) Crich,
D.; Yao, Q. J. Org. Chem. 1995, 60, 84. (c) Parker, K. A.; Spero, D. M.; Inman, K. C. TetrahedronLett. 1986, 27, 2833.
21
undergo a 5-endo cyclization, followed by hydrogen abstraction to form the desired
compound 30, or it can undergo an unusual 4-exo cyclization to form the 4-membered
heterocycle 38. A β-scission of this species to give radical 39, followed by 5-endo
cyclization and hydrogen abstraction, would afford isomer 31. The 5-endo and 4-exo
radical cyclizations are rarely seen.65 However, the products from 5-endo and 4-exo
cyclization would both be stabilized; the former by interaction with the sulfur atom
and the latter by being tertiary. This probably allows radical cyclization to compete
more favorably with hydrogen abstraction.
38
TBDMSO
S
TBDMSO
S
39 31
3037
β-scission2) [H]
1) 5-endoTBDMSO
S
4-exo
[H]5-endo TBDMSO
S
TBDMSO
S
TBDMSO
S
Scheme 15. Proposed mechanisms for the formation of compounds 30 and 31 from radical 37.
The sulfide 41, unsubstituted in the allylic moiety, was also prepared in order to study
the outcome of radical cyclization (Scheme 16). Lithiation of bromide 24, followed by
sulfur insertion and alkylation with allyl bromide, afforded the allylic sulfide 40 in a
43 % isolated yield from the bromide 24. Bromination followed by regeneration of the
double bond by treatment with NaBH4/bis(2-thienyl) ditelluride66 furnished the allylic
sulfide 41 in a moderate yield. Under standard radical cyclization conditions, 5-[(tert-
butyldimethylsilyl)oxy]-3,4,6,7-tetramethyl-2,3-dihydrobenzo[b]thiophene (42) was
obtained as the sole product in a 55 % isolated yield. As expected, 5-exo cyclization is
65 (a) Fossey, J.; Lefort, D.; Sorba, J. Free Radicals in Organic Chemistry, John Wiley & Sons, New
York, 1995. For some examples of 5-endo-trig and 4-exo-trig radical cyclizations, see: (b) Journet,M.; Rouillard, A.; Cai, D.; Larsen, R. D. J. Org. Chem. 1997, 62, 8630. (c) Ikeda, M.; Ohtani, S.;Yamamoto, T.; Sato, T.; Ishibashi, H. J. Chem. Soc., Perkin Trans. 1 1998, 1763. (d) Ishibashi, H.;Higuchi, M.; Ohba, M.; Ikeda, M. Tetrahedron Lett. 1998, 39, 75. (e) Bogen, S.; Gulea, M.;Fensterbank, L.; Malacria, M. J. Org. Chem. 1999, 64, 4920.
66 Engman, L. Tetrahedron Lett. 1982, 23, 3601.
22
much more rapid than 6-endo cyclization in the case of an unsubstituted allylic
moiety.67
TBDMSO
Br
TBDMSO
S
TBDMSO
S
Br TBDMSO
S
a,b,c d,e
f
24 40
41 42
Scheme 16. (a) t-BuLi, THF, -78 °C. (b) Sulfur, THF, -78 °C. (c) Allyl bromide, THF, -78 °C � 20 °C,43 % from 24. (d) Br2, AcOH, 20 °C. (e) NaBH4, bis(2-thienyl) ditelluride, EtOH, 50 % from 40. (f) n-Bu3SnH, AIBN, C6H6, reflux, 55 %.
An analogue of the vinylic sulfide 32, unsubstituted in the aromatic ring, has been
reported to rearrange to 3,3-dimethyl-2,3-dihydrobenzo[b]thiophene in
dichloromethane in the presence of excess aluminium chloride.68 The analogous
treatment of compound 32, resulted in desilylation but no cyclization occured.
A palladium catalyzed tandem cyclization-anion capture process69 was also attempted
for the cyclization. However, treatment of compound 29 (Scheme 12) with
palladium(II) acetate, triphenylphosphine, sodium formate, and tetraethylammonium
chloride in acetonitrile failed to give any product. This may be explained by catalyst
poisoning caused by the sulfur. Recently, 3,3-dimethyl-2,3-dihydrobenzo[b]thiophene
was prepared by Crich47f via an intramolecular homolytic substitution at sulfur and by
Hillhouse70 via thiametallacycles. However, similar methods were not investigated for
the preparation of compound 30.
2.2.2 Towards selenium and tellurium analogues of vitamin EII
Introduction
67 Abeywickrema, A. N.; Beckwith, A. L. J. J. Chem. Soc., Chem. Commun. 1986, 464.68 Derouane, D.; Harvey, J. N.; Viehe, H. G. J. Chem. Soc., Chem. Commun. 1995, 993.69 Burns, B.; Grigg, R.; Santhakumar, V.; Sridharan, V.; Stevenson, P.; Worakun, T. Tetrahedron
1992, 48, 7297.
23
For the preparation of 3,3,4,6,7-pentamethyl-2,3-dihydrobenzo[b]selenophene-5-ol
(21b), it was decided to try a radical cyclization approach analogous to the method
described for the synthesis of the sulfur analogue 30. The first simplified target
molecule was the selenide 44. Its attempted preparation is outlined in Scheme 17.
Treatment of compound 24 with tert-butyl lithium, elemental selenium followed by
addition of iodoethane in a one-pot procedure, gave the desired selenide 43 in a 76 %
isolated yield. However, bromination using bromine in dichloromethane, chloroform,
pentane or toluene at ambient temperature or reflux did not result in any formation of
compound 44. Instead, formation of diorganylselenium dibromide or organylselenium
tribromide products was observed. A different approach was therefore considered.
4324
da,b,cTBDMSO
SeCH2CH3
TBDMSO
Br
TBDMSO
SeCH2CH3
Br
44
Scheme 17. (a) t-BuLi, THF, -78 °C. (b) Selenium, THF, -78 °C. (c) Iodoethane, THF, -78 °C � 20 °C,76 % from 24. (d) Bromination under various conditions.
Except for the intramolecular homolytic substitution at selenium and tellurium, there
are few ways described in the literature to construct the skeletons of 2,3-
dihydrobenzo[b]selenophenes and tellurophenes and their respective 3,3-dimethyl
analogues.71 In order to be able to study a series of simple chalcogen analogues of
vitamin E, it was decided to try to construct the series 2,3-dihydrobenzo[b]furan-5-ol
and its 1-thio, 1-seleno and 1-telluro analogues (45a-d) using intramolecular
homolytic substitution at the chalcogen.
70 Han, R.; Hillhouse, G. L. J. Am. Chem. Soc. 1998, 120, 7657.71 For some examples, see (a) Renson, M. Chem. Scr. 1975, 8A, 29. (b) Hanold, N.; Meier, H. Chem.
Ber. 1985, 118, 198. (c) Ref. 70.
24
X
HO
45a X=O45b X=S45c X=Se45d X=Te
Schiesser and co-workers have recently paid considerable attention to intramolecular
homolytic substitution at selenium for the preparation of a wide variety of selenium-
containing heterocycles.48 Recently, the synthesis of benzo[b]selenophenes via
intramolecular homolytic substitution at selenium followed by aromatization was
reported.48b,d It was shown that the combination of an iodide radical precursor and a
benzyl leaving group was ideal for ring closure using standard radical techniques (a
�metal hydride� and AIBN in refluxing benzene). For example, the iodide 46a
afforded an excellent yield of benzo[b]selenophene 47 upon treatment with tris-
(trimethylsilyl)silane under standard radical conditions, whereas the bromide 46b gave
mainly the selenosilane 48 under identical conditions (Scheme 18).
Br
SeSi(SiMe3)3R OH
X
SeCH2PhR OH
a(X=Br)
a(X=I)
46a X=I46b X=Br
Se
R
4748
Scheme 18. (a) (Me3Si)3SiH, AIBN, C6H6.
These observations are in accord with experimental studies,72 as well as ab initio
calculations,46 suggesting that the rate of homolytic substitution at halogen and
chalcogen in a given row in the periodic table are very similar and it increases as one
traverses the groups of halogen/chalcogen. However, both rate constants and
computational data suggest that iodide precursors of type 49 may only be expected to
give about 50 % of the cyclized and aromatized telluride 50 together with the same
quantity of the undesired tellurosilane 51.
72 (a) Curran, D. P.; Martin-Esker, A. A.; Ko, S.-B.; Newcomb, M. J. Org. Chem. 1993, 58, 4691. (b)
Curran, D. P.; Bosch, E.; Kaplan, J.; Newcomb, M. J. Org. Chem. 1989, 54, 1826.
25
TeR'R OH
I
aTeSi(SiMe3)3
R OH
ITe
R
49 50
+
51
Scheme 19. (a) (Me3Si)3SiH, AIBN, C6H6.
Serendipity in action
Initially, it was decided to prepare 3-methyl benzo[b]tellurophene (55) in an
analogous way to that used for benzo[b]selenophenes.48b,d The aim was to prepare
telluride 52 by treatment of epoxide 5348d with a suitable ditelluride under reductive
conditions. Initially, the di-n-butyl ditelluride was tried because it is readily prepared
from n-butyllithium and elemental tellurium.73 In addition, the n-butyl radical is a
poorer leaving radical than the benzyl radical. This may direct the attack of the silyl
radical on iodine rather than on tellurium. Accordingly, the oxirane 5348d was reacted
with 2 equivalents of sodium n-butyltellurolate in THF/methanol (Scheme 20).74 To
our surprise, none of the desired telluride 52 was formed. Instead the major product
was the cyclized telluride 54, which was isolated in a 62 % yield. In the presence of a
catalytic amount p-toluenesulfonic acid, compound 54 could then be readily
dehydrated to give the tellurophene 55.75
I
TeOH
aO
I
aTe
OH
bTe
5552 53 54
Scheme 20. (a) di-n-butyl ditelluride, NaBH4, THF and MeOH. (b) p-TsOH, C6H6.
The mechanism for the cyclization is probably an alkyltelluride-mediated SRN1/SHi
tandem reaction according to Scheme 21.49 It is believed that the telluride 52 is rapidly
formed by nucleophilic substitution. This is then one-electron reduced via a
73 Cava, M. P.; Engman, L. Synth. Commun. 1982, 12, 163.74 Lucas, M. A.; Schiesser, C. H. J. Org. Chem. 1996, 61, 5754.75 Drakenberg, T.; Hörnfeldt, A.-B.; Gronowitz, S.; Talbot, J.-M.; Piette, J.-L. Chem. Scr. 1978-1979,
13, 152.
26
butyltelluride-mediated SRN1 mechanism76 to give the aryl radical 56 after loss of
iodide. The aryl radical can then undergo an intramolecular homolytic substitution at
tellurium to form the desired cyclized telluride 54 and an n-butyl radical. The radical
chain can then propagate with the n-dibutyl telluride radical anion serving as the
chain-transfer agent. This reaction represents, to the best of our knowledge, the first
example of an intramolecular homolytic substitution at tellurium. Interestingly,
addition of more than 2 equivalents of the tellurolate gave no improvement in yield,
whereas 1 equivalent resulted in a lower yield of compound 54. These observations
are in accord with the mechanism proposed in Scheme 21, as 2 equivalents of n-
butyltellurolate is required to generate 1 equivalent of compound 54 together with 1
equivalent of di-n-butyl telluride.
Te
OH
54
BuTeBu
OHTeBu
I52
OHTeBu
I
Bu
OHTeBu
BuTe
BuTeBu
I
56
Initiation: 52BuTe
52
Scheme 21. Proposed mechanism for the formation of the cyclized telluride 54
via an alkyl-telluride mediated SRN1/SHi reaction.
76 (a) Bunnett, J. F. Acc. Chem. Res. 1978, 11, 413. (b) Rossi, R. A.; de Rossi, R. H. Aromatic
Substitution by the SRN1 Mechanism; American Chemical Society:Washington, DC, 1983. Forrecent examples, see (c) Nazareno, M. A.; Rossi, R. A. J. Org. Chem. 1996, 61, 1645. (d)Beugelmans, R.; Chbani, M.; Soufiaoui, M. Tetrahedron Lett. 1996, 37, 1603. (e) Uneyama, K.;Yanagiguchi, Y.; Asai, H. Tetrahedron Lett. 1997, 38, 7763.
27
In order to prove the involvement of aryl radicals, other iodides were examined
(Scheme 22). Treatment of the selenium-containing iodide 5748d with 1 equivalent of
n-butyltellurolate in THF/MeOH for 24 h gave the cyclized selenide 58 in a 74 %
isolated yield. However, the reaction of the iodide 5977 was much slower under these
conditions and was found to be only 50 % complete after 24 h. Due to the instability
of the telluride 60 it could only be isolated in a yield of 17 % after flash
chromatography. Crich and co-workers recently reported similar stability problems
with some structurally related tellurides.78 Notably, Beckwith and Palacios have
reported similar intramolecular homolytic addition chemistry using PhS- and Ph2P-.79
Similar chemistry using alkyne 61 gave a mixture of compounds, where the product
62 was observable by 1H NMR spectroscopy. However, its lability precluded its
isolation and characterization. The formation of cyclized products in these reactions
leaves no doubt about the involvement of aryl radicals. In addition, rate data provided
further evidence for a radical pathway for the chemistry described (vide infra).
I
SeCH2PhOH
a
Se
OH
I
O
a(17 %)
(74 %)
O
Te
I
O
a
O
Te
57 58
59 60
61 62
Scheme 22. (a) di-n-butyl ditelluride, NaBH4, THF and MeOH.
The substituent on the tellurolate
After having established that n-butyltellurolate can be used for the preparation of
seleno- and tellurophenes, it was of interest to study the role of the alkyl group on
77 Goering, H. L.; Jacobson, R. R. J. Am. Chem. Soc. 1958, 80, 3277.78 Crich, D.; Chen, C.; Hwang, J.-T.; Yuan, H.; Papadatos, A.; Walter, R. I. J. Am. Chem. Soc. 1994,
116, 8937.79 Beckwith, A. L. J.; Palacios, S. M. J. Phys. Org. Chem. 1991, 4, 404.
28
tellurium. Would the alkyl group play an important role in the overall transformation
by changing the electron transfer properties of the tellurolate and the ease in which the
aryl radical undergoes homolytic substitution? In order to answer these questions, the
effect of other tellurolates was also investigated. Thus, the oxirane 53 was reacted
with both s- and t-butyltellurolate. Whereas the s-butyltellurolate was prepared in the
same manner as n-butyltellurolate, t-butyltellurolate was generated in situ from t-
butyllithium and elemental tellurium.14d In order to exclude the involvement of
borohydride in the chemistry described, n-butyltellurolate was also prepared
analogously to t-butyltellurolate. Reactions involving s- and t-butyltellurolate
appeared to proceed less cleanly than the reaction involving n-butyltellurolate (38 and
43 % yields, respectively for compound 54). The reaction with lithium n-
butyltellurolate gave 50 % yield of the desired compound 54. Accordingly, the method
of tellurolate generation does not seem to be a critical factor. It is concluded from
these experiments that borohydride probably does not play an important role in the
SRN1 process and that alkyl substitution has no significant effect either on the electron
transfer properties of the tellurolate or the intramolecular homolytic substitution
chemistry at tellurium.
Phenyltellurolate generated from both ditelluride by sodium borohydride and in situ
from phenyllithium and elemental tellurium, afforded only starting material and
diphenyl ditelluride. It is possible that the substrate undergoes rapid electron transfer
from PhTe-, and the resulting phenyltellanyl radical then undergoes rapid
recombination to afford diphenyl ditelluride.
Other tellurophenes
Having investigated the role of the substituent on the tellurolate, it was of similar
interest to see if the substituent on the oxirane plays any significant role in the
chemistry described. It was decided to try to synthesize benzo[b]tellurophene (66) and
3-phenyl benzo[b]tellurophene (67). When oxirane 63a,48d which is unsubstituted on
the oxirane moiety, was treated with n-butyltellurolate, an inseparable mixture of the
uncyclized alcohol 64 and the desired cyclized telluride 65 were obtained in a 1:1 ratio
(Scheme 23). The mixture was then treated with p-toluenesulfonic acid in refluxing
benzene to obtain the tellurophene 66 free from the alcohol 64 after flash
chromatography. A probable explanation for the formation of the alcohol 64 would be
29
an attack by the tellurolate at the benzylic position followed by a reduction by the
borohydride. This has previously been observed in the analogous selenium
chemistry.48d When compound 63b, with a phenyl group in the oxirane moiety, was
treated with n-butyltellurolate, none of the expected cyclized telluride 67 was
observed. Instead, the aldehyde 68 was isolated in a 45% yield. This compound is
probably formed via a 1,2-hydride shift as depicted in Scheme 23. This can be
explained by the lability of the oxirane, which has been shown to rearrange
quantitatively to aldehyde 68 upon prolonged storage at ambient temperature.
I
OR
BuTe-
(R=H)
OH
I Te
OH
+ H+
Te
63a R=H63b R=Ph
BuTe-
(R=Ph)
[1,2] hydride shift CHO
I
PhPh
HO-
HI
64 65 66
67 68Te
Ph
(R=Ph)BuTe-
Scheme 23
Rate constant for intramolecular homolytic substitution at tellurium
In addition to the cyclized telluride 54, a small amount of the reduced compound 70
was also isolated in a 7% yield in the reaction of oxirane 53 with n-butyltellurolate
(Scheme 24). The product distribution can be used to estimate, for the first time, the
rate of intramolecular homolytic substitution at tellurium. The integrated rate
expression, equation 4,48d which integrates to equation 5 under pseudo first order
conditions in THF can be used.
d[54] / d[70] = kC / (kH [THF]) (4)
[54] / [70] = kC / (kH [THF]) (5)
30
If one assumes that the concentration of THF is 12 M,80 that [54] / [70]= 57/7= 8.14
and that the rate constant for hydrogen abstraction of the aryl radical from THF (kH) is
5.1x106 M-1s-1 at 25 °C,81 the rate constant for cyclization of 69 (kC) is estimated to be
approximately 5x108 s-1 at 25 °C.
TeOHR
kC
54Te
OH
69
kHTe
OHR
H70
O
I53
n-BuTe
Scheme 24
This value can be compared with that of intramolecular homolytic substitution at
selenium in selenide 71 (expulsion of a benzyl radical), which has been estimated to
be approximately 3x107 s-1 at 80 °C.48d The larger rate constant of the telluride was
expected on the basis of the trends observed in the rate constants for intermolecular
homolytic substitution at selenium and tellurium and the relative leaving group
abilities.46
SeCH2Ph
71
Synthesis of 2,3-dihydrobenzo[b]selenophene-5-ol and its 1-telluro analogue
Having demonstrated that the alkyltelluride mediated SRN1/SHi tandem reaction can be
used for the preparations of selenophenes and tellurophenes, this methodology was
further employed in the synthesis of the selenium and tellurium-containing vitamin E
analogues 2,3-dihydrobenzo[b]selenophene-5-ol (45c) and 2,3-dihydro-
benzo[b]tellurophene-5-ol (45d). The compounds were prepared in several steps
starting from the commercially available 3-hydroxyphenylacetic acid (72), as outlined
80 The concentration of THF is approximately 12 M based on a density of 0.886 (25 °C).81 Landolt-Börnstein, Numerical Data and Functional Relationships in Science and Technology;
Fischer, H., Ed.; Springer-Verlag: Berlin, 1994; Vol. II/18.
31
in Scheme 25 and 26. After esterification of the starting material,82 the phenol was
protected as a TBDMS ether 73 in good yield. Reduction of this material with lithium
aluminium hydride afforded the primary alcohol 74 in excellent yield. Aromatic
iodination with iodine and silver trifluoroacetate gave, regioselectively, the iodide 75
in almost quantitative yield. The alcohol function was further converted, via the
mesylate, to afford the key intermediate tert-butyldimethylsilyl 4-iodo-3-(2-
iodoethyl)phenyl ether (76) in good yield. Treatment of the diiodide 76 with
NaBH4/Bn2Se283 afforded the benzyl selenide 77 in an isolated yield of 84 %. The
alkyltelluride mediated SRN1/SHi reaction described above could then be used to
induce ring closure. Thus, treatment of the benzyl selenide with one equivalent of n-
butyltellurolate (NaBH4/di-n-butyl ditelluride73) afforded the cyclized selenide 78 in a
65 % isolated yield.
HOCO2H TBDMSO
CO2Et TBDMSO OH
TBDMSO OH
I
TBDMSO I
I
TBDMSO SeCH2Ph
I Se
TBDMSO
Se
HO
a,b c
d e,f g
h or i j
72 73 74
7675
77 78 45c
Scheme 25. (a) EtOH, H2SO4, reflux, 94 %. (b) TBDMSCl, imidazole, DMF, 20 °C, 98 %. (c) LiAlH4,Et2O, 0 °C, 96 %. (d) I2, CF3CO2Ag, 20 °C, 96 %. (e) MsCl, Et3N, CH2Cl2, 0 °C, quantitative. (f) NaI,acetone, reflux, 88 %. (g) Bn2Se2, NaBH4, EtOH, 20 °C, 84 %. (h) n-Bu2Te2, NaBH4, THF, MeOH, 20°C, 65 %. (i) TTMSS, AIBN, benzene, reflux, 48 %. (j) TBAF, THF, 20 °C, 95 %.
In a similar way, treatment of the diiodide 76 with two equivalents of n-
butyltellurolate afforded the desired cyclized telluride 79 in an isolated yield of 47 %
(Scheme 26). The cyclized selenide 78 could also be obtained from iodide 77 by
silane mediated intramolecular homolytic substitution, but in a somewhat lower yield
82 Guanti, G.; Banfi, L.; Riva, R. Tetrahedron 1994, 50, 11945.
32
(48 %). Finally, the cyclized selenide and telluride were deprotected with TBAF to
obtain the desired products 45c and 45d in isolated yields of 95 % and 65 %,
respectively.
TBDMSO I
I
a b
76 79 45dTe
TBDMSO
Te
HO
Scheme 26. (a) n-Bu2Te2, NaBH4, THF, MeOH, 20 °C, 47 %. (b) TBAF, THF, 20 °C, 65 %.
2.2.3. Synthesis of 2,3-dihydrobenzo[b]furan-5-ol and its 1-thio analogueIII
In order to study the antioxidant properties of the series of unmethylated chalcogen
compounds 45a-d, the oxygen and sulfur derivatives 2,3-dihydrobenzo[b]furan-5-ol
(45a) and 2,3-dihydrobenzo[b]thiophene-5-ol (45b) were prepared. The oxygen
analogue was synthesized according to a literature procedure.84 The sulfur analogue
has been previously prepared,85 but an alternative procedure involving intramolecular
homolytic substitution at sulfur has been developed (Scheme 27). The ester 73 was
treated with bromine in acetic acid to obtain aryl bromide 80 in a high yield. The ester
group was then reduced with lithium aluminium hydride, and the resulting alcohol 81
was treated with diphenyl disulfide and tri-n-butylphosphine in benzene to obtain
sulfide 82 in a 76 % isolated yield. The sulfide together with tri-n-butyltin hydride and
AIBN in refluxing benzene afforded the cyclized sulfide 83, via intramolecular
homolytic substitution at sulfur. Due to an inseparable impurity, the sulfide 83 was
deprotected with TBAF without isolation to give the desired product 45b in an
isolated yield of 62 % from compound 82.
83 Reich, H. J.; Jasperse, C. P.; Renga, J. M. J. Org. Chem. 1986, 51, 2981.84 Alabaster, R. J.; Cottrell, I. F.; Marley, H.; Wright, S. H. B. Synthesis 1988, 950.85 Zambias, R. A.; Hammond, M. L. Synth. Commun. 1991, 21, 959.
33
TBDMSOCO2Et
TBDMSOCO2Et
Br
TBDMSO
Br
OH
TBDMSO
Br
SPh
S
HO
73
S
TBDMSO
80 81
82 83 45b
a b
c d e
Scheme 27. (a) Br2, KOAc, HOAc, 15-17 oC, 97 %. (b) LiAlH4, Et2O, 0 oC, 93 %. (c) (PhS)2, n-Bu3P,C6H6, ambient temperature, 76 %. (d) n-Bu3SnH, AIBN, C6H6, reflux. (e) TBAF, THF, ambienttemperature, 62 % from 82.
2.3 The antioxidant profile of 2,3-Dihydrobenzo[b]furan-5-ol and its 1-thio, 1-
seleno and 1-telluro analoguesIII
2.3.1 Introduction
Prevention of autoxidation is an important task for antioxidants. Efficient phenolic
antioxidants should have either a weaker O-H bond than the C-H bond in the material
where oxidation should be prevented and/or the peroxyl radicals formed in the
propagation step must easily oxidize the antioxidants. Thus, the redox and
thermochemical properties of antioxidants are in crucial need of investigation.
Therefore, we have determined the bond dissociation enthalpies, the redox properties
and the hydrogen transfer capabilities for the series of antioxidants 45a-d. The
capability of the compounds to inhibit stimulated lipid peroxidation, to catalyze
decomposition of hydrogen peroxide in the presence of glutathione and their
inhibiting effect on peroxidation in liver microsomes were also investigated.
2.3.2 Redox and thermochemical properties
Pulse radiolysis is a powerful technique for the determination of one-electron
reduction potentials.86 This is the only method available for the measurement of
thermodynamically correct potentials of solvated radicals. The technique is based on
the formation of the radicals eaq-, H·, and HO· by radiolysis of water. These radicals
can in turn be transformed into well defined reducing or oxidizing radicals via the use
of suitable scavengers, such as N3-.
34
In order to determine the one-electron reduction potentials of the phenoxyl radicals,
corresponding to phenols 45a-d, primary oxidation of the phenolates ArO- was
achieved by azide radicals (N3·). The azide radical N3· was produced in the pulse
radiolysis experiment by the reaction of HO· with N3-. The reduction potential could
then be determined from the redox equilibrium between the radical of interest and a
redox couple with a known one-electron reduction potential (equation 6), using the
Nernst equation (∆E°= 0.0591logK at 298 K).
ArO + Ref ArO- + Ref (6)
Having determined the pKa values of the phenols spectrophotometrically, the O-H
bond dissociation enthalpies in solution could then be calculated from equation 7,
which is derived from Hess� law.87,88 The results are summarized in Table 1.
BDE(O-H)= 96.48E° + 5.70pKa + C (7)
From these data one can see that the pKa values as well as the reduction potentials do
not differ dramatically within the series of compounds (45a-d). Consequently, the
bond dissociation enthalpies are similar for all four compounds (336-340 kJmol-1).
These values are significantly higher than the value reported for α-tocopherol (323
kJmol-1).89
Table 1. Acidity and redox properties of compounds 45 in aqueous solution
Compound pKa(H2O)
λmax ArO·(nm)
Referencesubstance
K E° (ArO· / ArO-)(V vs NHE)
BDE O-H(kJ mol-1)
45a 10.6 450 45b 1 0.49 34045b 10.0 560 4-MeOC6H4OH 7.86 0.49 33745c 9.9 630 45a 1 0.49 33645d 9.5 800 4-MeOC6H4OH 2.53 0.52 337
Cyclic voltammetry is a versatile method for studying simple redox processes.86 The
cyclovoltammetric oxidation potentials for compounds 45a-d are given in Table 2.
86 Daasbjerg, K.; Pedersen, S. U.; Lund, H. In General Aspects of the Chemistry of Radicals, Alfassi,
Z. B., Ed., John Wiley & Sons, West Sussex, 1999; Chap. 12.87 Lind, J.; Shen, X.; Eriksen, T. E.; Merényi, G. J. Am. Chem. Soc. 1990, 112, 479.88 Jonsson, M.; Lind, J.; Eriksen, T. E.; Merényi, G. J. Chem. Soc., Perkin Trans. 2 1993, 1567.
35
The oxidation potential for the oxygen analogue (1.35 V vs NHE) is irreversible
whereas it is quasireversible for the sulfur analogue (1.35 V vs NHE). For the selenide
and telluride derivatives the oxidation potentials are fully reversible (1.13 and 0.74 V
vs NHE, respectively), indicating the formation of stable radical cations in these cases.
Table 2. Oxidation potentials as determined by cyclic voltammetry
Compound E (V vs Fc) E (V vs NHE) Solvent45a 0.66 1.35 MeCN45b 0.66 1.35 MeCN45c 0.44 1.13 MeCN45d 0.05 0.74 MeCN
2.3.3 Reactivity towards tert-butoxyl radicals
It is known that tert-butoxyl radicals rapidly abstract phenolic hydrogen atoms in
comparison with decomposition by β-scission.90 This can be used for studying the
hydrogen donating ability of antioxidants. As demonstrated by Ingold and co-workers,
such rapid processes are conveniently studied by laser flash photolysis.91
Photodecomposition of di-tert-butyl peroxide by 355-nm laser pulses gives rise to
tert-butoxyl radicals within the duration of the pulse. These radicals will rapidly react
with the antioxidants to generate phenoxyl radicals, which can be directly monitored
by UV-spectroscopy. The build-up of the signal follows pseudo-first order kinetics
(equation 8) where kobs is the observed rate constant and k is the absolute rate constant
for the reaction of the tert-butoxyl radical with the phenolic compound (equation 9).
Me3CO + ArOH (9)
(8)
k Me3COH + ArO
kobs = k0 + k[ArOH]
The pseudo first order rate constant kobs was determined from the build-up curves of
the respective phenoxyl radicals at their λmax values (Table 1). The absolute rate
constant k was then calculated by a least-square fitting of kobs versus [ArOH] for five
different concentrations. The rate constant values for the oxygen 45a and sulfur 45b
89 Wayner, D. D. M.; Lusztyk, E.; Ingold, K. U.; Mulder, P. J. Org. Chem. 1996, 61, 6430.90 Das, P. K.; Encinas, M. V.; Steenken, S.; Scaiano, J. C. J. Am. Chem. Soc. 1981, 103, 4162.91 (a) Foti, M.; Ingold, K. U.; Lusztyk, J. J. Am. Chem. Soc. 1994, 116, 9440. (b) Valgimigli, L.;
Banks, J. T.; Ingold, K. U.; Lusztyk, J. ibid. 1995, 117, 9966. (c) Valgimigli, L.; Ingold, K. U.;Lusztyk, J. ibid. 1996, 118, 3545.
36
analogues were both determined to be 2x108 M-1s-1. For the selenium and tellurium
analogues, the corresponding rate constants could not be determined due to laser
photo-excitation of the compounds. The rate constant for sulfide 21a was determined
to be 4x108 M-1s-1. As compared with the unmethylated sulfide 45b, the better
hydrogen donating ability of sulfide 21a is due to the more efficient stabilization of
the phenoxyl radical by the electron releasing substituents on the aromatic ring. For
comparison, the rate constant for α-tocopherol was determined to be 6x108 M-1s-1,
indicating that α-tocopherol is a somewhat better hydrogen donor compared to
compounds 45a, 45b, and 21a. With the limited rate data at hand, it seems that the
reactivity of the tert-butoxyl radicals with the series of compounds 45 is reflected by
the phenolic O-H bond dissociation enthalphies given in Table 1.
2.3.4 Inhibition of lipid peroxidation
Lipid peroxidation is a detrimental process, since it causes damage to lipids in
biological membranes. The prevention of lipid peroxidation is therefore an important
task for antioxidants. It has earlier been demonstrated that some organotellurium
compounds retard lipid peroxidation.21b-d In order to study the ability to inhibit lipid
peroxidation, azo-initiated peroxidation of linoleic acid or derivatives thereof have
been used.92 Recently a two-phase system for the assessment of antioxidant capacity
was reported (Figure 4a).21d The aqueous phase contains N-acetylcysteine (NAC) as
the stoichiometric reductant, which could regenerate any of the tellurium (IV)
compounds (Scheme 8) or selenium (IV) compounds formed. The organic phase
(chlorobenzene) contains linoleic acid, a radical initiator (AMVN) and the antioxidant
to be studied.
92 (a) Niki, E.; Saito, T.; Kawakami, A.; Kamiya, Y. J. Biol. Chem. 1984, 259, 4177. (b) Cosgrove, J.
P.; Church, D. F.; Pryor, W. A. Lipids 1987, 22, 299. (c) Braughler, J. M.; Pregenzer, J. F. FreeRad. Biol. Med. 1989, 7, 125. (d) Pryor, W. A.; Cornicelli, J. A.; Devall, L. J.; Tait, B.; Trivedi, B.K.; Witiak, D. T.; Wu, M. J. Org. Chem. 1993, 58, 3521.
H11C5(CH2)6CO2H
CN
N N
NC
Linoleic acid
AMVN
Antioxidant to be studied
+
+
H2O / NAC
Chlorobenzene
αααα -tocopherol
0
100
200
300
400
500
600
0 0,5 1 1,5 2 2,5Time (h)
[LO
OH
] (M
)
37
a b
Figure 4. a) Two-phase system for determination of antioxidant capacity. b) Concentration/time plot asexemplified with α-tocopherol (LOOH = linoleic acid hydroperoxide).
The lipid peroxidation was initiated by the addition of the radical initiator and the
solution was stirred vigorously at 42 °C. During the peroxidation, linoleic acid is
converted to a conjugated lipid hydroperoxide,3 which can be monitored by HPLC
with UV detection at 234 nm. A typical concentration-time plot for α-tocopherol is
depicted in Figure 4b. Two parameters can be determined from the trace: the inhibited
rate of peroxidation (Rinh) and the inhibition time (Tinh). The inhibited rate, Rinh, is a
measure of antioxidant efficiency, which is determined by least-square methods from
the curve during the inhibited phase of peroxidation. The inhibition time, Tinh, is a
measure of the antioxidant�s lifetime and is determined graphically as the cross-point
for the inhibited and uninhibited lines.
Values of Rinh and Tinh for compounds 45a-d are shown in Table 3 together with those
recorded for α-tocopherol, sulfide 21a and telluride 20. For comparison, the values
were also determined in the absence of a reducing agent.
It is clear from the inhibited rates of peroxidation that α-tocopherol is more efficient
than any of the analogues tested, both in the presence and in the absence of N-
acetylcysteine. In the absence of N-acetylcysteine the tellurides 45d and 20 do not
inhibit peroxidation at all and the selenide 45c does so poorly. The tellurides 45d and
20 are probably oxidized immediately by the residual hydroperoxide contained in the
linoleic acid, and thus inactivated. The oxygen and sulfur analogues 45a and 45b,
have a longer induction time than α-tocopherol, but they are less efficient as
antioxidants. When a reductant is present in the aqueous phase, the antioxidant
efficiency increases in the series of compounds 45a-d as one traverse the periodic
38
table from oxygen to tellurium. The short inhibition time of telluride 45d can
probably be explained by the decomposition of the telluride (or its corresponding
telluroxide), since increasing the amount of thiol in the aqueous phase could not
prolong the inhibition time. In the presence of NAC, all the inhibition times are
increased, except for α-tocopherol and the sulfide 21a. This seems to indicate that the
thiol is capable not only of regenerating selenides and tellurides from their oxides, but
also some of the phenols at the aqueous lipid interphase. The latter process is
apparently much less efficient for sterically hindered phenols. However, the presence
of methyl groups in the sulfide 21a significantly increases antioxidant efficiency as
compared with the unmethylated sulfide 45b. This was also reflected in their rate
constants for hydrogen atom donation to tert-butoxyl radicals (vide supra).
In summary, it was demonstrated that the selenide 45b and the tellurides 45d and 20
could be regenerated in the presence of a stoichiometric reductant and thereby been
able to act as catalytic antioxidants.
Table 3. Inhibited rate of peroxidation, Rinh, and time of inhibition, Tinh, for antioxidants tested inthe presence and absence of N-acetylcysteine (NAC)
Antioxidant Rinh and Tinh in the presence of NAC Rinh and Tinh in the absence of NACRinh (µM/h)a Tinh (min)b Rinh (µM/h)a Tinh (min)b
45a 44 140 46 9045b 36 180 30 8545c 36 >300 64 5045d 17 60 370 0α-Tocopherol 11 90 10 80Sulfide 21a 21 70 20 65Telluride 20 34 160 370 0a Rate of peroxidation during the inhibited phase (uninhibited rate ~370 µM/h).b Duration of the inhibited phase of peroxidation.
The chain-breaking properties of phenolic aryl alkyl tellurides can be ascribed to their
ability to reduce peroxyl radicals. In doing so, the diorganyl telluride will be oxidized
to a tellurium (IV) dihydroxide. As illustrated in Scheme 28 for the telluride 45d
several mechanisms must be considered for this transformation. Phenolic diorganyl
tellurides can donate a hydrogen atom to form a resonance-stabilized phenoxyl
radical. Alternatively, there could also be an electron transfer from tellurium to give a
radical cation. The formation of this species will result in an increase in the acidity of
the phenolic hydrogen, which will readily be lost to give the resonance stabilized
39
phenoxyl radical. Oxidation and hydrolysis of the resulting phenoxyl radical then
affords the thiol reducible tellurium (IV) dihydroxide 84. Considering the observed
reversible oxidation of the telluride 45d at 0.74 V vs NHE (vide supra), the chain-
breaking mechanism probably involves an electron transfer from tellurium. Similar
mechanisms are likely to be operative for the selenide 45c as well. A conclusion
regarding the true chain-breaking mechanism has to await further mechanistic
investigations.
Te
HO
Te
HO
OHHO
O
Te
Te
O
Hydrogen atom/electron-proton transfer:
1) Oxidation2) H2O
-H or
-e / -H
45d84
Scheme 28. Proposed chain-breaking mechanisms for telluride 45d.
2.3.5 Hydroperoxide decomposing capacity
In biological systems the selenium-containing glutathione peroxidases (GSH-px)
catalyze�s the reduction of various hydroperoxides by reduced glutathione (eqn 10).
The catalytic cycle relies on the redox cycling of selenium. Recently, much attention
has been paid to the development of compounds that mimic the action of GSH-px.
The selenium-containing heterocycle Ebselen (85) was the first compound found to
exhibit this property.93
SeN
O
85
In addition to Ebselen, a wide variety of selenium-containing compounds have been
found to catalyze the reduction of hydroperoxides in the presence of thiols.93 It has
been demonstrated that diorganyl tellurides and ditellurides are also glutathione
93 Mugesh, G.; Singh, H. B. Chem. Soc. Rev. 2000, 29, 347 and references cited therein.
40
peroxidase mimics.21c-d,f The catalytic mechanism for diaryl tellurides21d and other
organotelluriums94 rely on redox cycling at tellurium (Scheme 8).
The thiol peroxidase activity for the series of unmethylated vitamin E analogues 45a-
d was assessed using a coupled reductase assay (equations 10 and 11).21a
2 GSH + ROOHGSH-px or mimic
GSSG + ROH + H2O (10)
GSSGGSSG-reductase
NADPH2 GSH (11)
In this assay, hydroperoxide, glutathione and the antioxidant to be evaluated are
allowed to react at pH 7.4 in the presence of glutathione reductase and NADPH.
When the glutathione has been oxidized to the corresponding disulfide, it will be
enzymatically reduced by GSSG-reductase with NADPH acting as a co-factor and
stoichiometric reductant. The reaction was followed spectrophotometrically by
observing the consumption of NADPH at 340 nm and the catalytic activity was
determined by recording the initial rate increase in comparison with the uncatalyzed
process. The results are shown in Table 4. As expected, the oxygen and sulfur
analogues 45a-b showed essentially no catalytic activity. The selenide 45c also did not
show any catalytic activity, whereas the telluride 45d was a highly active catalyst.
When hydrogen peroxide was used as the oxidant, NADPH was consumed almost 100
times faster than in the control experiment. To show that the telluride 45d acts in a
catalytic fashion under the conditions used, additional NADPH, H2O2 and glutathione
were added to the incubation mixture and the consumption of NADPH was recorded
again. After five repeated additions, no significant change in catalytic efficiency was
observed. When tert-butyl hydroperoxide and cumene hydroperoxide were used as
oxidants, the reaction was accelerated even more (333 and 213 times, respectively, as
compared to the uncatalyzed reaction). This makes the telluride 45d one of the most
efficient glutathione peroxidase mimics described.
Table 4. Glutathione peroxidase-like activity of compounds 45a-d with various hydroperoxidesubstrates
Catalyst Hydroperoxide NADPH consumptiona (µM/min) % Catalysisb
- H2O2 6.3±0.1 -
94 Detty, M. R.; Friedman, A. E.; Oseroff, A. R. J. Org. Chem. 1994, 59, 8245.
41
45a H2O2 6.1±0.3 9745b H2O2 6.0±0.2 9545c H2O2 6.5±0.3 10345d H2O2 624±56c 9900- t-BuOOH 1.1±0.2 -45c t-BuOOH 1.4±0.2 12745d t-BuOOH 365±17 33300- CumylOOHd 2.4±0.4 -45c CumylOOHd 3.7±0.4 15645d CumylOOHd 511±26 21300a Calculated over a 20 s period of stable decline of absorption at 340 nm.b The catalyst percentage increase of the basal reaction rate between GSH and H2O2 was calculated asrate of NADPH consumption + 5 mol % catalyst (µM/min) / rate of NADPH consumption + vehicle(µM/min) x 100.c The following values [µM NADPH/min (% catalysis)] were recorded after repeated additions of 250µM NADPH, 250 µM H2O2 and 500 µM GSH: 628.1 (9968), 640.2 (10161), 635.3 (10079), 622.8(9885), 612.9 (9728).d The incubation contained 20 % DMSO.
2.3.6 Lipid peroxidation in liver microsomes
Lipid peroxidation is accelerated in the presence of Fe(II) and Fe(III) salts and a
reducing agent such as ascorbate.95 Thus, microsomal fractions prepared from animal
tissues undergo lipid peroxidation in the presence of these substances. The metal
dependent decomposition of lipid hydroperoxides results in the formation of carbonyl-
containing products, some of which can react with thiobarbituric acid to give
thiobarbituric reactive substances (TBARS) that can easily be quantified by UV-
spectroscopy at 532 nm.96 By measuring the amount of TBARS formed, one can
determine how efficient the antioxidants are to inhibit stimulated lipid peroxidation.
This is expressed as the IC50 value, which is the concentration that inhibits 50 % of
the TBARS formation in the control experiment.
It has previously been demonstrated that diaryl tellurides retard lipid peroxidation in
liver microsomes.21b The oxygen, sulfur and selenium compounds 45a-c showed only
weak inhibiting activity (IC50-values of ca 250, 25, 13 µM, respectively), whereas the
organotellurium compound 45d was a potent inhibitor of lipid peroxidation in liver
microsomes (IC50-value of 0.13 µM). The outstanding performance of the telluride
indicates that peroxide-decomposing capacity may be more important than the chain-
breaking ability to inhibit lipid peroxidation in microsomes under the conditions used.
95 Sevanian, A.; Nordenbrand, K.; Kim, E.; Ernster, L.; Hochstein, P. Free Rad. Biol. Med. 1990, 8,
145.96 Greenwald, R. A., Ed., CRC Handbook of Methods for Oxygen Radical Research, CRC Press: Boca
Raton, 1985.
42
Ascorbate may also serve to amplify the protective effect by continuously regenerating
the organotellurium compound from its corresponding oxidation products.
2.4 Towards a selenium-containing tocopherolIV
2.4.1 Introduction
Although Ingold showed that 1-thio-α-tocopherol (16e) and the related 6-
hydroxythiochromanes are slightly less efficient antioxidants than their corresponding
oxygen analogues,55d-f we believe that a selenium or tellurium containing α-tocopherol
would show very interesting antioxidative properties as a consequence of their ability
to undergo redox cycling. Initially, it was decided to develop a procedure for the
synthesis of the selenium-containing tocopherol analogue 2-methyl-2-(4,8,12-
trimethyltridecyl)-selenochroman-6-ol (86).
Se
HO
86
2.4.2 Strategy
The crucial step in the preparation of compound 86 is the construction of the
selenium- containing six-membered ring with a quaternary carbon next to selenium. A
good way to do this is probably via an intramolecular homolytic substitution at the
heteroatom. In order to investigate this methodology, we decided to use a method for
radical generation developed by Barton, Crich, and co-workers97 in combination with
intramolecular homolytic substitution at selenium (Figure 5). Tertiary radicals can be
generated from tertiary alcohols by the use of PTOC oxalates (87) in the following
way. Treatment of a tertiary alcohol with excess oxalyl chloride in benzene would
convert the alcohol into the corresponding oxalyl monochloride alkyl ester (88). The
crude material is then added to a suspension of N-hydroxypyridine-2-thione sodium
salt and DMAP in benzene to obtain N-(alkoxyoxalyloxy)pyridine-2-thione (87),
which readily undergoes decomposition on heating to generate a tertiary radical. It
was envisaged that a free radical chain reaction would then produce the desired
43
selenide via an intramolecular homolytic substitution at selenium with the expulsion
of a benzyl-leaving group (Figure 5, lower part).
SeRR'
SeCH2Ph
OR
R'
OO
O
N
S
SeCH2Ph
OR
R'
ClO
O
SeCH2Ph
OHR
R'
+N SCH2PhSeCH2Ph
RR'
88
87
SeCH2Ph
OR
R'
OO
O
N
S
87
Initiation:
N S+87
SeCH2Ph
RR'
Propagation:
∆
C6H6 DMAPC6H6
(COCl)2
+ 2 CO2
SHi
+ 2 CO2
CH2Ph
N SO-Na+
Figure 5. Strategy for the construction of a six-membered selenium-containing heterocycle with aquaternary carbon-center α to the selenium atom.
2.4.3 Synthesis of the model compound 2,2-dibutyl selenacyclohexane
To probe the versatility of the proposed strategy, the model compound 2,2-dibutyl
selenacyclohexane (92) was synthesized from commercially available methyl 5-
bromovalerate (89) as depicted in Scheme 29. Treatment of the bromide 89 with
sodium benzylselenolate afforded the selenide 90 in high yield. The required tertiary
alcohol 91 was then prepared in a 76 % yield by treatment of the selenide 90 with two
equivalents of n-butylmagnesium bromide. The alcohol 91 was then treated with
97 (a) Barton, D. H. R.; Crich, D. J. Chem. Soc., Chem. Commun. 1984, 774. (b) Barton, D. H. R.;
Crich, D. Tetrahedron Lett. 1985, 26, 757. (c) Barton, D. H. R.; Crich, D. J. Chem. Soc., PerkinTrans. 1 1986, 1603. (d) Crich, D.; Fortt, S. M. Synthesis 1987, 35.
44
oxalyl chloride and pyridine, followed by the sodium salt of 2-mercaptopyridine N-
oxide and DMAP to obtain the desired selenide 92 in a yield of 40 % from compound
91. This is, to our knowledge, the first example where a tertiary radical undergoes an
intramolecular homolytic substitution at selenium to generate a selenium-containing
heterocycle. Attempted intramolecular homolytic substitution using the tertiary iodide
93 under various standard radical conditions [tris(trimethylsilyl)silane and AIBN, n-
Bu3SnH and AIBN, (n-Bu3Sn)2 and hν], gave only low yields of the desired selenide
92. Elimination of hydriodic acid seems to be a severe competing reaction in these
cases. The possibility of preparing the telluride 96 was also investigated. Thus,
treatment of the ester 89 with n-butyltellurolate gave the telluride 94 in good yield.
This compound was then treated with n-butylmagnesium bromide to afford the desired
tertiary alcohol 95 in a 59 % isolated yield. However, treatment of the alcohol 95
under the Barton/Crich conditions did not result in any product formation. Thus, the
methodology seems to be limited to the preparation of selenium-containing
heterocycles.
a b
89 90 R=Bn, X=Se (94%)94 R=Bu, X=Te (87%)
c,dX
HO Bu
BuR
91 R=Bn, X=Se (76%)95 R=Bu, X=Te (59%)
92 X=Se (40%)96 X=Te (no product)
XCO2CH3RBr
CO2CH3
XBuBu
Scheme 29. (a) Bn2Se2 or n-Bu2Te2, NaBH4, EtOH. (b) Mg, n-BuBr, Et2O. (c) Oxalyl chloride,pyridine, C6H6. (d) Sodium salt of 2-mercaptopyridine N-oxide, DMAP, C6H6.
I Bu
BuPhCH2Se93
2.4.4 Synthesis of a selenium-containing tocopherol analogue
The synthesis of 2-methyl-2-(4,8,12-trimethyltridecyl)-selenochroman-6-ol (86) was
accomplished as outlined in Schemes 30 and 31. Bromination of the bromide 98
45
(obtained from alcohol 97 by a literature procedure98) gave the dibromide 99 in high
yield. Alkylation of the enolate of tert-butyl acetoacetate with this bromide gave the
ester 100. Hydrolysis and decarboxylation furnished the desired ketone 101
quantitatively. The carbonyl functionality was protected as a 1,3-dioxalane (102). This
material was lithiated, elemental selenium inserted and the resulting selenolate
alkylated with benzyl bromide in a one-pot procedure. After workup, the crude
material was subjected to deprotection with hydrochloric acid to furnish the key
intermediate 2-(2-benzylselenenyl-5-methoxyphenyl)ethyl methyl ketone (103) in a 78
% isolated yield from compound 102.
CH3OOH
CH3OBr
CH3O
Br
O
CO2tBu
CH3O
Br
O
a b c
CH3O
SeCH2Ph
Bu
HO
Se
HO
Bu
97 98 99
101
CH3O
Br
OO
CH3O
SeCH2Ph
O
CH3O
Br
Br
d,e f g,h,i,j
k l,m,n
102
103
100
104 105
Scheme 30. (a) PBr3, Et2O, 94 %. (b) Br2, CHCl3, 97%. (c) tert-Butyl acetoacetate, NaH, THF, 71 %.(d) HCl (6M). (e) ∆, quantitative from 100. (f) Ethylene glycol, p-TsOH, C6H6, 97%. (g) t-BuLi, THF.(h) Elemental Se. (i) Benzyl bromide. (j) HCl (3M), 78% from 102. (k) Mg, n-BuBr, Et2O, 74%. (l)Oxalyl chloride, pyridine, C6H6. (m) Sodium salt of 2-mercaptopyridine N-oxide, DMAP, C6H6. (n)BBr3, CH2Cl2, 28 % from 104.
In order to test the methodology, an n-butyl group was first introduced as the
lipophilic tail. Thus, treatment of the ketone 103 with n-butylmagnesium bromide
afforded the tertiary alcohol 104 in an isolated yield of 74%. The alcohol was then
treated with oxalyl chloride and pyridine, followed by the sodium salt of 2-
mercaptopyridine N-oxide and DMAP to obtain the desired cyclized selenide together
with an inseparable impurity. The crude selenide was subjected to deprotection using
98 Beard, Jr, W. Q.; van Eenam, D. N.; Hauser, C. R. J. Org. Chem. 1961, 26, 2310.
46
boron tribromide to furnish the desired selenochromane 105 in a 28 % isolated yield
from compound 104.
Encouraged by this result, it was decided to introduce the racemic vitamin E side-
chain. In order to do so, the bromide 106 was prepared in six steps from commercially
available farnesol, according to literature procedures.99
Br
106
The Grignard reagent of the bromide 106 was treated with ketone 103 to afford the
tertiary alcohol 107 in a 54 % isolated yield (Scheme 31). This compound was then
subjected to the Barton/Crich procedure and deprotection, as described for the
preparation of the selenide 105, to afford the final selenochromane 86 in a yield of 19
% from the compound 107 as a mixture of diastereomers.
CH3O
SeCH2Ph
HOH
103 a b,c,d
Se
HO
H
107 86
3
3
Scheme 31. (a) Mg, bromide 106, Et2O, 54 %. (b) Oxalyl chloride, pyridine, C6H6. (c) Sodium salt of2-mercaptopyridine N-oxide, DMAP, C6H6. (d) BBr3, CH2Cl2, 19 % from 107.
2.5 Conclusions and outlook
The fully methylated sulfide 21a was prepared by both ionic and radical chemistry and
interesting rearrangements were observed in the two synthetic pathways.
A new methodology for the synthesis of dihydroselenophene and dihydrotellurophene
derivatives is described. In these reactions a tellurium-mediated tandem SRN1/SHi
sequence was suggested to be operative. The rate constant for intramolecular
homolytic substitution at tellurium was also determined to be 5x108 s-1 at 25 °C.
A series of unmethylated antioxidants 45a-d were prepared in order to study their
antioxidant capacity as a function of the chalcogen atom. The sulfur analogue 45b was
99 (a) Aldrich, J. R.; Oliver, J. E.; Lusby, W. R.; Kochansky, J. P.; Borges, M. J. Chem. Ecol. 1994,
20, 1103. (b) Cohen, N.; Schaer, B. J. Org. Chem. 1992, 57, 5783.
47
prepared by intramolecular homolytic substitution at sulfur. For the corresponding
selenium and tellurium analogues (45c-d), intramolecular homolytic substitution via
the SRN1/SHi mechanism was used. Pulse radiolysis experiments showed that the
reduction potentials of the compounds as well as the O-H bond dissociation enthalpies
were essentially independent of the chalcogen atom. In a two-phase lipid peroxidation
model it was demonstrated that the selenium and tellurium containing antioxidants
45c-d could be regenerated in the presence of a stoichiometric reductant. As expected,
introduction of methyl groups into the sulfur compound accelerated its hydrogen
donating ability and thereby its antioxidant capacity. We also demonstrated that the
tellurium-containing antioxidant 45d not only acted as a very efficient glutathione
peroxidase mimic, but also as an efficient inhibitor of lipid peroxidation in liver
microsomes.
A novel methodology for the synthesis of the selenium-containing vitamin E analogue
86 was also developed. The crucial step relied on formation of a tertiary radical from a
PTOC oxalate in a system where intramolecular homolytic substitution could occur at
selenium. Unfortunately, this methodology was limited to selenium-containing
heterocycles. In the future it is hoped to develop procedures that can be used for
making the corresponding tellurium analogue. It could also be worthwhile to prepare
the fully methylated selenium and tellurium-containing analogues 21b-c by the
SRN1/SHi tandem reaction developed. Provided that these materials show interesting
antioxidative properties, the ultimate goal would then be to synthesize the �real�
selenium and tellurium analogues of α-tocopherol.
48
3. Organotellurium compounds as stabilizers for polymericmaterialsV
3.1 Introduction
Inorganic tellurium compounds (e.g. elemental tellurium and its oxides and halides)
have been shown to stabilize various polymers.100 However, organotellurium
compounds have not been extensively studied for this purpose. It has previously been
demonstrated that diaryl tellurides could stabilize natural rubber, polyethylene and
polypropylene.20 As mentioned in Section 1.2.3, organic tellurides can act both as
chain-breaking antioxidants and as peroxide decomposers. It has also been shown that
diaryl tellurides act as quenchers of singlet oxygen (1O2),101 which is a contributor to
the photo-oxidative degradation of various polymers.102 These results prompted us to
carry out a more thorough investigation of the stabilizing effect of diaryl tellurides in
polymers. Because of its thermal stability (up to 220°C) and its stabilizing effects in
natural rubber at the 1-2% level,20 bis[4-(dimethylamino)phenyl] telluride (108) has
been selected for these studies.
2
NMe
MeTe
108
The polymer chosen for this investigation was the thermoplastic elastomer PACREL ,
which is composed of a continuous polypropylene phase together with a lightly
crosslinked polybutylacrylate phase.103 PACREL is prepared by a solid-state
polymerization technique, which has been recently patented.104 Because of their
100 (a) Abed-Ali, S. S.; McWhinnie, W. R.; Scott, G. Polym. Degrad. Stabil. 1988, 21, 211. (b)
Sawada, Y.; Zen, S.; Toyoshima, T.; Konno, T. Jpn. Pat. 04,216,852 (1992). (c) Newton, H. V. U.S.Pat. 4,265,800 (1981). (d) Cain, M. E.; Knight, G. T.; Gazeley, K. F.; McHugh, P. L. Ger. Pat.2,121,747 (1971). (e) Chimura, I.; Ito, K.; Ishii, Y.; Watanabe, M.; Mori, M.; Fuchinabe, T. Jpn.Pat. 7,105,249 (1967).
101 Serguievski, P.; Detty, M. R. Organometallics 1997, 16, 4386.102 Wiles, D. M. In Singlet Oxygen. Reaction with Organic Compounds and Polymers, Rånby, B. and
Rabek, J. F., Eds., John Wiley & Sons: New York, 1978; Chap. 34.103 Vainio, T.; Jukarainen, H.; Seppälä, J. J. Appl. Polym. Sci. 1996, 59, 2095.104 Vestberg, T.; Lehtiniemi, I. U.S. Pat. 5,164,456 (1992), U.S. Pat. 5,300,578 (1994), U.S. Pat.
5,312,872 (1994).
49
interesting thermal characteristics, there is an enormous interest in thermoplastic
elastomers in industry today.105
3.2 Preparation of bis[4-(dimethylamino)phenyl] telluride
The diaryl telluride bis[4-(dimethylamino)phenyl] telluride (108) was prepared
according to a literature procedure (Scheme 32).106 The complex 109 was prepared by
the addition of aniline to a suspension of TeCl4 in ether. This material was then
reduced with Na2S2O5 and treated with NaHCO3 to afford a mixture of the telluride
108 and its corresponding ditelluride. Treatment with activated copper in 1,4-dioxane
effected detelluration of the ditelluride to give the desired monotelluride 108 in a fair
yield.
2
NMe
MeTe
108
c
ba 109
108 NMe
MeTe
2
+
NMe
Me2 2:1-Complex+ TeCl4
Scheme 32. (a) Et2O, ambient temp., 92 %. (b) Na2S2O5, CH2Cl2, H2O followed by NaHCO3 untilneutral. (c) Activated copper (Cu(0)), 1,4-dioxane, reflux, 58 % from complex 109.
3.3 Autoxidation and mechanical properties in polymers
3.3.1 Chemiluminescence
The oxidation of most organic materials is accompanied by low intensity emission of
light. This is called chemiluminescence (CL) if the light is the result of a chemical
reaction. Russell has formulated the predominating mechanism behind CL, as outlined
in Scheme 33.107 The low intensity emission of light is due to relaxation of excited
molecules to their ground state. Emission of light from polymers has been studied
since 1961,108 but not until recently has it been possible to measure it with high
105 Reish, M. S. C&EN 1996 (August 5), 10.106 Engman, L.; Persson J. Organometallics 1993, 12, 1068.107 (a) Russell, G. A. J. Am. Chem. Soc. 1957, 79, 3871. (b) Howard, J. A.; Ingold, K. U. J. Am. Chem.
Soc. 1968, 90, 1056.108 Ashby, G. E. J. Polym. Sci. 1961, 50, 99.
50
accuracy. Nowadays, it is possible to detect a few photons per second. Therefore, the
CL technique is a very sensitive method that can be used to study kinetics of polymer
oxidation.109
RO2
*
R' R'
O + hνννν
R OO O
O
R'R'
HR' R'
OO
+- ROH- O2
R'
O
R'
Scheme 33. The Russell mechanism for chemiluminescence
3.3.2 Mechanical properties
There are various methods available to study the mechanical properties of
polymers.110 One way is to determine the modulus, which is defined as the ratio of the
stress applied to a material to the strain produced. Tensile strength is another
commonly used quantity for evaluating the strength of polymers. The tensile strength
is a measure of the tension required for the material to break. Elongation at break, in
percent, is also a measure of polymer strength. Since polymers may show anisotropy
due to high shear rate during the shaping process, it may be relevant to determine
these properties in two different directions, both in the flow (or machine) direction
(MD) and in the transverse direction (TD). Dynamic mechanical tests represent other
methods for evaluating the mechanical properties of polymers. These tests measure
the response of a material to a sinusoidal stress. From these measurements one
obtain�s a damping term, tan δ, which is a measure of the mechanical losses in the
material, originating from e.g. movements of chain segments in the polymer. Tan δ is
invaluable in the examination of relaxation processes and transitions in polymers.
109 (a) Mendenhall, G. D. Angew. Chem., Int. Ed. Engl. 1990, 29, 362. (b) Forsström, D.; Kron, A.;
Stenberg, B.; Terselius, B; Reitberger, T. Polym. Degrad. Stabil. 1994, 43, 277. (c) Mattson, B.;Reitberger, T.; Stenberg, B.; Terselius, B. Polym. Test. 1991, 10, 399 and references cited therein.(d) Jipa, S.; Setnescu, R.; Setnescu, T.; Zaharescu, T. Polym. Degrad. Stab. 2000, 68, 165.
110 Nielsen, L. E.; Landel, R. F. Mechanical Properties of Polymers and Composites, 2nd ed., MarcelDekker: New York, 1994.
51
3.4 Results and discussion
3.4.1 Mechanical properties
Mechanical properties for different PACREL materials were studied with and
without the addition of the diaryl telluride 108. The results are summarized in Tables
5-8.
Table 5. Mechanical properties of PACREL���� 206 blends
Unstabilized material Stabilized material (0.3 % telluride)Aging time (weeks) 0 4 4 0 4 4Aging temp. (°C) 23 23 40 23 23 40Hardness (°/shore A) 74 75 76 74 74 76Modulus (MPa) MD 11 ± 0.5 11 ± 0.3 12 ± 0.5 9 ± 0.3 9 ± 0.3 10 ±0.2 TD 8 ± 0.4 8 ± 0.2 8 ± 0.5 8 ± 1 8 ± 1 9 ± 0.3Tensile strength (MPa) MD 4 ± 0.1 4 ± 0.1 4 ± 0.1 4 ± 0.1 4 ± 0.1 4 ± 0.1 TD 3 ± 0.1 3 ± 0.1 3 ± 0.2 4 ± 0.2 4 ± 0.1 4 ± 0.1Elongation at break (%) MD 139 ± 7 147 ± 6 108 ± 8 185 ± 6 185 ± 4 181 ± 9 TD 221 ± 23 222 ± 24 222 ± 31 259 ± 19 258 ± 5 246 ± 12MD, Machine direction; TD, Transverse direction.
In Table 5 the mechanical properties of PACREL 206 blends are shown. It was
observed that the hardness of the polymer was unaffected by the addition of 0.3 % of
the telluride 108 whereas the tensile modulus in the machine direction, but not in the
transverse direction, was slightly lowered. Other interesting results are as follows: the
tensile strength in the transverse direction was improved by 25 % by the addition of
the stabilizer whereas the elongation at break was increased by about 11-17 %.
Notably, the improvements in mechanical properties were already apparent before the
aging of the polymer.
The results with PACREL 632 are given in Table 6. With this polymer a lower
amount of stabilizer was used (0.2 %), and the aging time was prolonged. The aging
temperature was also increased to 80 °C as compared to 40 °C in Table 5. Again, the
hardness of the polymer was unaffected by aging with the presence of the telluride
108. For the tensile modulus no trends were observed. Regarding the tensile strength
in the machine direction, the stabilized material was outstanding in comparison with
the unstabilized material after aging for 10 weeks at 80 °C. The tensile strength in the
machine direction for the unstabilized material decreased by approximately 50
%,
52
53
whereas the tensile strength for the telluride stabilized material remained unchanged
throughout the aging process.
There was also a significant difference between stabilized and unstabilized material
after 10 weeks of aging at 80 °C for the elongation at break in the machine direction.
In this case the stabilized material was unaffected after aging whereas the values for
the unstabilized polymer were only about 10 % of those recorded for the stabilized
material.
In Table 7 the corresponding results are summarized for PACREL 632 blends
compounded with oil and commercial stabilizers. Because this polymer contains oil,
no direct comparison can be made with the results in Tables 5 and 6. However, it
should be clear that the telluride increases the resistance to aging of PACREL at least
as much as conventional stabilizers.
Table 7. Mechanical properties of PACREL���� 632 blends compounded with oil and conventionalantioxidants. No telluride present.
Aging time (weeks) 0 10 4 10Aging temp. (°C) 23 23 80 80Hardness (°/shore A) 85 86 - 82Modulus (MPa) MD 69 ± 13 46 ± 3 65 ± 10 41 ± 5 TD 34 ± 3 30 ± 3 35 ± 4 33 ± 3Modulus 100% (MPa) MD 7 ± 0.3 8 ± 0.1 8 ± 0.3 8 ± 0.3 TD 5 ± 0.1 5 ± 0.2 5 ± 0.1 6 ± 0.2Tensile Strength (MPa) MD 8 ± 0.4 8 ± 0.1 8 ± 0.3 8 ± 0.3 TD 7 ± 0.6 7 ± 0.3 7 ± 0.5 8 ± 0.5Elongation at break (%) MD 133 ± 5 130 ± 1 109 ± 4 110 ± 6 TD 188 ± 36 216 ± 9 196 ± 21 198 ± 20MD, Machine direction; TD, Transverse direction.
In another series of experiments, the amount of stabilizer and the aging times were
varied for PACREL 631 . The results are summarized in Table 8. These experiments
were also complemented with chemiluminescence measurements (vide infra). Again,
one could observe an improvement in the mechanical properties of the stabilized
material as compared to the unstabilized. In the case of the stabilized polymer (0.17 %
telluride), both the tensile strength and the elongation at break were almost unchanged
54
after 9 weeks of aging at 70 °C. However, for the unstabilized material the
corresponding values were reduced, the former by 40 % and the latter by 75 %.
From the above results there is no doubt that PACREL needs to be stabilized in
some way to withstand thermo-oxidative degradation. As compared to the
unstabilized material, the diaryl telluride 108 (0.17-0.50 %) considerably improved
elongation at break and tensile strength of the unaged as well as aged polymer.
However, it is still not obvious why the diaryl telluride causes such an improvement in
the mechanical properties of the unaged material. The known hydroperoxide
decomposing capacity of the diaryl telluride 108 could be one contributing
explanation to the observed effect.
Table 8. Mechanical properties of PACREL���� 631 blends
Telluride 108(%)
Aging Temp(°C)
Aging Time(weeks)
Tensile Strength(MPa)
Elongation at Break(%)
0 23 0 3.8 ± 0.2 79 ± 220 70 4 2.6 ± 0.5 25 ± 90 70 9 2.3 ± 0.3 21 ± 4
0.17 23 0 7.1 ± 0.5 1240.17 70 4 6.4 ± 0.8 84 ± 340.17 70 9 7.4 ± 0.6 130 ± 450.30 23 0 6.2 ± 0.1 197 ± 80.30 70 4 6.6 ± 0.5 168 ± 290.30 70 9 6.5 ± 0.7 146 ± 300.50 23 0 7.1 ± 0.7 200 ± 680.50 70 4 7.4 ± 0.3 167 ± 250.50 70 9 7.6 ± 0.5 184 ± 25
3.4.2 Chemiluminescence
Chemiluminescence-time traces recorded in oxygen at 150 °C for PACREL 631
stabilized with 0.17-0.50 % of the diaryl telluride 108 showed significantly longer
induction times (60-70 min) than the unstabilized material (30 min). This indicates
that the antioxidative properties of tellurides previously observed in organic solvents
can also be expressed in a polymeric matrix.
In another experiment the total luminiscence intensity (TLI) was measured as a
function of the aging time at 70 °C (Figure 6) where PACREL 631 was stabilized
with 0, 0.17, 0.30, and 0.50 % of the telluride. From these measurements one can
observe that the initially low TLI for stabilized materials is only slightly increased (2-
5 times) by aging. However, for the unstabilized material, the initial TLI is
55
approximately 10 times higher than that for the stabilized material and it increases by
a factor of about 13 during aging. These studies clearly show that the diaryl telluride
108 considerably prolongs the induction period of oxygen-stimulated oxidation of the
polymer and drastically reduces the total luminescence intensity of unaged and aged
samples.
Figure 6. TLI as a function of aging time at 70 °C for samples of PACREL 631 stabilized with 0,0.17, 0.30, 0.50 % diaryl telluride 108. The right luminescence scale is for unstabilized material only.
3.4.3 Dynamic mechanical properties
The dynamic mechanical properties were studied as a function of temperature for the
unstabilized and stabilized (0.3 % of the telluride 108) PACREL 631 . The results are
shown in Figure 7 as a plot of tan δ versus temperature.
56
Figure 7. Tan δ as a function of temperature for PACREL 631 stabilized with 0 % (normal line) and0.30 % (bold line) of the diaryl telluride 108.
This figure clearly shows that the addition of the stabilizer suppresses the tan δ of the
polybutylacrylate phase, resulting in a decreased flexibility of the polymer containing
the telluride. However, the tensile modulus and the hardness are essentially unaffected
by admixing of the telluride and swelling experiments gave no indication of increased
crosslinking.
3.4.4 Stabilizer stability
The thermal stability of the diaryl telluride 108 was previously studied by
thermogravimetric analysis.20 These measurements indicated that 5 % of the initial
sample weight was lost at 220 °C in an atmosphere of nitrogen. To study the stability
of the telluride under more userlike conditions, neat telluride was heated in air at
various temperatures and 1H-NMR spectra were recorded. Heating of the telluride at
180 °C for 10 minutes gave a very slight decomposition of the material. At 215 °C,
about 20 % of the telluride was decomposed within 15 minutes. So far, the
decomposition products have not been further analyzed. However, no telluroxide
corresponding to telluride 108 was formed under the conditions used.
3.5 Conclusions and outlook
It has been demonstrated, by different techniques, that bis[4-(dimethylamino)phenyl]
telluride (108) acts as a multifunctional antioxidant/stabilizer in the thermoplastic
elastomer PACREL . Addition of the telluride 108 improved both the tensile strength
and elongation at break of the polymer. Chemiluminescence measurements
demonstrated that the telluride prolonged the induction period of thermo-oxidation
and reduced the total luminescence intensity of the material.
This project will now be expanded to involve other tellurides as well as other
polymers. It is also hoped that it will be possible to use diaryl tellurides in a catalytic
fashion in polymers, according to Scheme 8.
It is also planned to use these compounds for the stabilization of other synthetic and
man-made products such as oils, fluids and paper products.
57
5. Acknowledgements
I would like to express my sincere gratitude to:
My supervisor Prof. Lars Engman for accepting me as a graduate student, and for hisgreat support, patience and enthusiasm during these years.
Past and present members of group Engman:Dr. Vijay Gupta for your friendship, for taking care of me when I came to Uppsala andMelbourne and for sharing your knowledge in chalcogen and radical chemistry.Fil. Lic. Magnus Beşev for your friendship, your eagerness to discuss chemistry andfor introducing me to Raki.Dr. Takahiro Kanda, Stefan Berlin, Cecilia Ericsson, Dr. Andrei Vasiliev, Fil. Lic. PerEriksson, Dr. Nicolai Stuhr-Hansen, Dr. Nawaf Al-Maharik and Fredrik Lehmann fortheir friendship and stimulating co-operation.
Prof. Carl H. Schiesser for allowing me to work in his research group for three monthsand for making my stay at the University of Melbourne very pleasant.
Assoc. Prof. Mats Jonsson, Assoc. Prof. Ian A. Cotgreave, Assoc. Prof. LeifHammarström and Martin Sjödin for stimulating collaborations concerning thevitamin E project.
Prof. Bengt Stenberg, Dr. Martin Bellander, Dr. Karin Jacobsson and Dr. VivecaLönnberg, for fruitful collaborations concerning the polymer project.
Dr Nicholas Power and Ulrika Yngve for revision and linguistic corrections of thisthesis.
Assoc. Prof. Adolf Gogoll for keeping all the instruments and computers in goodshape and for his invaluable help concerning NMR and handling of instruments.
The technical and administrative staff Leffe, Gunnar, Wicke, Thomas and Tatti.
My first labmates Anders Persson and Andreas Palmgren, who took care of me when Istarted.Per Ryberg for friendship and our hard games of badminton.Stefan Modin for all help concerning HPLC, computers, etc.
58
All other past and present members of the department, especially Micke, Kattis, Sofia,Markus, Lotta, Catrin, Viviane, Tobias, Benita, Katarina, Pedro, Magnus E., Ludvig,Kalle and Jocke, for a pleasant time inside as well as outside Chemicum.
The Schiesser group and other people at the University of Melbourne who made mystay there great fun and memorable. I will never forget Melbourne and our trip to theGreat Ocean road. A special thanks goes to Drs Melissa Laws and Lisa Zugaro forfruitful collaborations.
The Swedish Research Council for Engineering Sciences, Stiftelsen Bengt Lundqvistminne, Knut och Alice Wallenbergs stipendiefond, Stiftelsen Sigurd och Elsa Goljesminne and C. F. Liljewalchs resestipendiefond for financial support.
My friends from the south, especially Patrik, Fredrik, Christian, Staffan, Mats, Manne,Lasse, Magnus W. and Stefan for always being there.
My family for their great support.
My beloved Ulrika for all your love, your encouragement and for making my life moreeventful and fun.